id
float64 2.21k
2.31k
| content
stringlengths 78
1.92k
⌀ | category
stringclasses 8
values |
|---|---|---|
2,312.06122 |
Caution: This paper includes offensive words that could potentially cause unpleasantness. The fast-paced evolution of generative language models such as GPT-4 has demonstrated outstanding results in various NLP generation tasks. However, due to the potential generation of offensive words related to race or gender, various Controllable Text Generation (CTG) methods have been proposed to mitigate the occurrence of harmful words. However, existing CTG methods not only reduce toxicity but also negatively impact several aspects of the language model's generation performance, including topic consistency, grammar, and perplexity. This paper explores the limitations of previous methods and introduces a novel solution in the form of a simple Gated Toxicity Avoidance (GTA) that can be applied to any CTG method. We also evaluate the effectiveness of the proposed GTA by comparing it with state-of-the-art CTG methods across various datasets. Our findings reveal that gated toxicity avoidance efficiently achieves comparable levels of toxicity reduction to the original CTG methods while preserving the generation performance of the language model.
|
cs
|
2,312.04956 |
Intrusion Detection Systems (IDS) are widely employed to detect and mitigate external network security events. VANETs (Vehicle ad-hoc Networks) are evolving, especially with the development of Connected Autonomous Vehicles (CAVs). So, it is crucial to assess how traditional IDS approaches can be utilised for emerging technologies. To address this concern, our work presents a stacked ensemble learning approach for IDS, which combines multiple machine learning algorithms to detect threats more effectively than single algorithm methods. Using the CICIDS2017 and the VeReMi benchmark data sets, we compare the performance of our approach with existing machine learning methods and find that it is more accurate at identifying threats. Our method also incorporates hyperparameter optimization and feature selection to improve its performance further. Overall, our results suggest that stacked ensemble learning is a promising technique for enhancing the effectiveness of IDS.
|
cs
|
2,312.03113 |
In GPU graph analytics, the use of external memory such as the host DRAM and solid-state drives is a cost-effective approach to processing large graphs beyond the capacity of the GPU onboard memory. This paper studies the use of Compute Express Link (CXL) memory as alternative external memory for GPU graph processing in order to see if this emerging memory expansion technology enables graph processing that is as fast as using the host DRAM. Through analysis and evaluation using FPGA prototypes, we show that representative GPU graph traversal algorithms involving fine-grained random access can tolerate an external memory latency of up to a few microseconds introduced by the CXL interface as well as by the underlying memory devices. This insight indicates that microsecond-latency flash memory may be used as CXL memory devices to realize even more cost-effective GPU graph processing while still achieving performance close to using the host DRAM.
|
cs
|
2,312.0218 |
We demonstrate for the first time that Galactic cosmic rays with energies as high as 1e10 eV can trigger a cascade of low-energy (< 20 eV) secondary electrons that could be a significant contributor to the interstellar synthesis of prebiotic molecules whose delivery by comets, meteorites, and interplanetary dust particles may have kick-started life on Earth. We explore the relative importance of low-energy (< 20 eV) secondary electrons--agents of radiation chemistry--and low-energy (< 10 eV), non-ionizing photons--instigators of photochemistry. Our calculations indicate fluxes of 100 electrons/cm2/s for low-energy secondary electrons produced within interstellar ices due to incident attenuated Galactic cosmic-ray (CR) protons. Consequently, in certain star-forming regions where internal high-energy radiation sources produce ionization rates that are observed to be a thousand times greater than the typical interstellar Galactic ionization rate, the flux of low-energy secondary electrons should far exceed that of non-ionizing photons. Because reaction cross-sections can be several orders of magnitude larger for electrons than for photons, even in the absence of such enhancement our calculations indicate that secondary low-energy electrons are at least as significant as low-energy (< 10 eV) non-ionizing photons in the interstellar synthesis of prebiotic molecules. Most importantly, our results demonstrate the pressing need for explicitly incorporating low-energy electrons in current and future astrochemical simulations of cosmic ices. Such models are critically important for interpreting James Webb Space Telescope infrared measurements, which are currently being used to probe the origins of life by studying complex organic molecules found in ices near star-forming regions.
|
astro-ph
|
2,312.04539 |
Vision-Language Models (VLMs) have emerged as promising tools for open-ended image understanding tasks, including open vocabulary segmentation. Yet, direct application of such VLMs to segmentation is non-trivial, since VLMs are trained with image-text pairs and naturally lack pixel-level granularity. Recent works have made advancements in bridging this gap, often by leveraging the shared image-text space in which the image and a provided text prompt are represented. In this paper, we challenge the capabilities of VLMs further and tackle open-vocabulary segmentation without the need for any textual input. To this end, we propose a novel Self-Guided Semantic Segmentation (Self-Seg) framework. Self-Seg is capable of automatically detecting relevant class names from clustered BLIP embeddings and using these for accurate semantic segmentation. In addition, we propose an LLM-based Open-Vocabulary Evaluator (LOVE) to effectively assess predicted open-vocabulary class names. We achieve state-of-the-art results on Pascal VOC, ADE20K and CityScapes for open-vocabulary segmentation without given class names, as well as competitive performance with methods where class names are given. All code and data will be released.
|
cs
|
2,312.05923 |
Video Individual Counting (VIC) aims to predict the number of unique individuals in a single video. % Existing methods learn representations based on trajectory labels for individuals, which are annotation-expensive. % To provide a more realistic reflection of the underlying practical challenge, we introduce a weakly supervised VIC task, wherein trajectory labels are not provided. Instead, two types of labels are provided to indicate traffic entering the field of view (inflow) and leaving the field view (outflow). % We also propose the first solution as a baseline that formulates the task as a weakly supervised contrastive learning problem under group-level matching. In doing so, we devise an end-to-end trainable soft contrastive loss to drive the network to distinguish inflow, outflow, and the remaining. % To facilitate future study in this direction, we generate annotations from the existing VIC datasets SenseCrowd and CroHD and also build a new dataset, UAVVIC. % Extensive results show that our baseline weakly supervised method outperforms supervised methods, and thus, little information is lost in the transition to the more practically relevant weakly supervised task. The code and trained model will be public at \href{}{CGNet}
|
cs
|
2,312.0369 |
Vitrimer is a new class of sustainable polymers with the ability of self-healing through rearrangement of dynamic covalent adaptive networks. However, a limited choice of constituent molecules restricts their property space, prohibiting full realization of their potential applications. Through a combination of molecular dynamics (MD) simulations and machine learning (ML), particularly a novel graph variational autoencoder (VAE) model, we establish a method for generating novel vitrimers and guide their inverse design based on desired glass transition temperature (Tg). We build the first vitrimer dataset of one million and calculate Tg on 8,424 of them by high-throughput MD simulations calibrated by a Gaussian process model. The proposed VAE employs dual graph encoders and a latent dimension overlapping scheme which allows for individual representation of multi-component vitrimers. By constructing a continuous latent space containing necessary information of vitrimers, we demonstrate high accuracy and efficiency of our framework in discovering novel vitrimers with desirable Tg beyond the training regime. The proposed vitrimers with reasonable synthesizability cover a wide range of Tg and broaden the potential widespread usage of vitrimeric materials.
|
cs
|
2,312.031 |
Quantum Key Distribution (QKD) enables two parties to establish a common secret key that is information-theoretically secure by transmitting random bits that are encoded as qubits and sent over a quantum channel, followed by classical information processing steps known as information reconciliation and key extraction. Transmission of information over a quantum channel introduces errors that are generally considered to be due to the adversary's tempering with the quantum channel and needs to be corrected using classical communication over an (authenticated) public channel. Commonly used error-correcting codes in the context of QKD include cascade codes, low-density parity check (LDPC) codes, and more recently polar codes. In this work, we explore the applicability of designing of a polar code encoder based on a channel reliability sequence. We show that the reliability sequence can be derived and used to design an encoder independent of the choice of decoder. We then implement our design and evaluate its performance against previous implementations of polar code encoders for QKD as well as other typical error-correcting codes. A key advantage of our approach is the modular design which decouples the encoder and decoder design and allows independent optimization of each. Our work leads to more versatile polar code-based error reconciliation in QKD systems that would result in deployment in a broader range of scenarios.
|
cs
|
2,312.06201 |
Carbon stars are excellent kinematic tracers of galaxies and play important roles in understanding the evolution of the Galaxy. Therefore, it is worthwhile to search for them in a large amount of spectra. In this work, we build a new carbon star catalog based on the LAMOST DR7 spectra. The catalog contains 4542 spectra of 3546 carbon stars, identified through line index and near-infrared color-color diagrams. Through visual inspection of the spectra, we further subclassify them into 925 C--H, 384 C--R, 608 C--N, and 1292 Ba stars. However, 437 stars could not be sub-classified due to their low signal-to-noise. Moreover, by comparing with LAMOST DR7 pipeline we find 567 more carbon stars and visually sub-classify them. We find that on the $J-H$ vs. $H-K_{\rm s}$ two-color diagram, C--N stars can be reliably distinguished from the other three sub-types. Additionally, by utilizing the Gaia distance, we study the distribution of carbon stars in the H-R diagram and identify 258 dwarf carbon stars by the criterion $M_{\rm G}>$5.0\,mag. Finally, we present the spatial distribution in Galactic coordinates of the 3546 carbon stars. The majority of C-N, C-R, and Ba stars are distributed at low Galactic latitudes, while most C--H and dC stars distribute at high Galactic latitudes.
|
astro-ph
|
2,312.05573 |
In the context of sketching for compressive mixture modeling, we revisit existing proofs of the Restricted Isometry Property of sketching operators with respect to certain mixtures models. After examining the shortcomings of existing guarantees, we propose an alternative analysis that circumvents the need to assume importance sampling when drawing random Fourier features to build random sketching operators. Our analysis is based on new deterministic bounds on the restricted isometry constant that depend solely on the set of frequencies used to define the sketching operator; then we leverage these bounds to establish concentration inequalities for random sketching operators that lead to the desired RIP guarantees. Our analysis also opens the door to theoretical guarantees for structured sketching with frequencies associated to fast random linear operators.
|
stat
|
2,312.0413 |
In this paper, we establish sharp dispersive estimates for the linear wave equation on the lattice $\mathbb{Z}^d$ with dimension $d=4$. Combining the singularity theory with results in uniform estimates of oscillatory integrals, we prove that the optimal time decay rate of the fundamental solution is of order $|t|^{-\frac{3}{2}}\log |t|$, which is the first extension of P. Schultz's results \cite{S98} in $d=2,3$ to the higher dimension. We also observe that the decay rate for $d=2,3,4$ can be well interpreted by the Newton polyhedra.
|
math
|
2,312.02239 |
Channel charting builds a map of the radio environment in an unsupervised way. The obtained chart locations can be seen as low-dimensional compressed versions of channel state information that can be used for a wide variety of applications, including beam prediction. In non-standalone or cell-free systems, chart locations computed at a given base station can be transmitted to several other base stations (possibly operating at different frequency bands) for them to predict which beams to use. This potentially yields a dramatic reduction of the overhead due to channel estimation or beam management, since only the base station performing charting requires channel state information, the others directly predicting the beam from the chart location. In this paper, advanced model-based neural network architectures are proposed for both channel charting and beam prediction. The proposed methods are assessed on realistic synthetic channels, yielding promising results.
|
cs
|
2,312.06249 |
We prove a structure theorem for ergodic homological rotation sets of homeomorphisms isotopic to the identity on a closed orientable hyperbolic surface: this set is made of a finite number of pieces that are either one-dimensional or almost convex. The latter ones give birth to horseshoes; in the case of a zero-entropy homeomorphism we show that there exists a geodesic lamination containing the directions in which generic orbits with respect to ergodic invariant probabilities turn around the surface under iterations of the homeomorphism. The proof is based on the idea of $\textit{geodesic tracking}$ of orbits that are typical for some invariant measure by geodesics on the surface, that allows to get links between the dynamics of such points and the one of the geodesic flow on some invariant subset of the unit tangent bundle of the surface.
|
math
|
2,312.05187 |
Large-scale automatic speech translation systems today lack key features that help machine-mediated communication feel seamless when compared to human-to-human dialogue. In this work, we introduce a family of models that enable end-to-end expressive and multilingual translations in a streaming fashion. First, we contribute an improved version of the massively multilingual and multimodal SeamlessM4T model-SeamlessM4T v2. This newer model, incorporating an updated UnitY2 framework, was trained on more low-resource language data. SeamlessM4T v2 provides the foundation on which our next two models are initiated. SeamlessExpressive enables translation that preserves vocal styles and prosody. Compared to previous efforts in expressive speech research, our work addresses certain underexplored aspects of prosody, such as speech rate and pauses, while also preserving the style of one's voice. As for SeamlessStreaming, our model leverages the Efficient Monotonic Multihead Attention mechanism to generate low-latency target translations without waiting for complete source utterances. As the first of its kind, SeamlessStreaming enables simultaneous speech-to-speech/text translation for multiple source and target languages. To ensure that our models can be used safely and responsibly, we implemented the first known red-teaming effort for multimodal machine translation, a system for the detection and mitigation of added toxicity, a systematic evaluation of gender bias, and an inaudible localized watermarking mechanism designed to dampen the impact of deepfakes. Consequently, we bring major components from SeamlessExpressive and SeamlessStreaming together to form Seamless, the first publicly available system that unlocks expressive cross-lingual communication in real-time. The contributions to this work are publicly released and accessible at
|
cs
|
2,312.04076 |
Low-shot image classification, where training images are limited or inaccessible, has benefited from recent progress on pre-trained vision-language (VL) models with strong generalizability, e.g. CLIP. Prompt learning methods built with VL models generate text features from the class names that only have confined class-specific information. Large Language Models (LLMs), with their vast encyclopedic knowledge, emerge as the complement. Thus, in this paper, we discuss the integration of LLMs to enhance pre-trained VL models, specifically on low-shot classification. However, the domain gap between language and vision blocks the direct application of LLMs. Thus, we propose LLaMP, Large Language Models as Prompt learners, that produces adaptive prompts for the CLIP text encoder, establishing it as the connecting bridge. Experiments show that, compared with other state-of-the-art prompt learning methods, LLaMP yields better performance on both zero-shot generalization and few-shot image classification, over a spectrum of 11 datasets.
|
cs
|
2,312.02431 |
In recent years, several machine learning models have been proposed. They are trained with a language modelling objective on large-scale text-only data. With such pretraining, they can achieve impressive results on many Natural Language Understanding and Generation tasks. However, many facets of meaning cannot be learned by ``listening to the radio" only. In the literature, many Vision+Language (V+L) tasks have been defined with the aim of creating models that can ground symbols in the visual modality. In this work, we provide a systematic literature review of several tasks and models proposed in the V+L field. We rely on Wittgenstein's idea of `language games' to categorise such tasks into 3 different families: 1) discriminative games, 2) generative games, and 3) interactive games. Our analysis of the literature provides evidence that future work should be focusing on interactive games where communication in Natural Language is important to resolve ambiguities about object referents and action plans and that physical embodiment is essential to understand the semantics of situations and events. Overall, these represent key requirements for developing grounded meanings in neural models.
|
cs
|
2,312.06309 |
In several branches of the social sciences and humanities, surveys based on standardized questionnaires are a prominent research tool. While there are a variety of ways to analyze the data, some standard procedures have become established. When those surveys want to analyze differences in the answer patterns of different groups (e.g., countries, gender, age, ...), these procedures can only be carried out in a meaningful way if there is measurement invariance, i.e., the measured construct has psychometric equivalence across groups. As recently raised as an open problem by Sauerwein et al. (2021), new evaluation methods that work in the absence of measurement invariance are needed.
This paper promotes an unsupervised learning-based approach to such research data by proposing a procedure that works in three phases: data preparation, clustering of questionnaires, and measuring similarity based on the obtained clustering and the properties of each group. We generate synthetic data in three data sets, which allows us to compare our approach with the PCA approach under measurement invariance and under violated measurement invariance. As a main result, we obtain that the approach provides a natural comparison between groups and a natural description of the response patterns of the groups. Moreover, it can be safely applied to a wide variety of data sets, even in the absence of measurement invariance. Finally, this approach allows us to translate (violations of) measurement invariance into a meaningful measure of similarity.
|
cs
|
2,312.06659 |
We establish the convergence of the unified two-timescale Reinforcement Learning (RL) algorithm presented by Angiuli et al. This algorithm provides solutions to Mean Field Game (MFG) or Mean Field Control (MFC) problems depending on the ratio of two learning rates, one for the value function and the other for the mean field term. We focus a setting with finite state and action spaces, discrete time and infinite horizon. The proof of convergence relies on a generalization of the two-timescale approach of Borkar. The accuracy of approximation to the true solutions depends on the smoothing of the policies. We then provide an numerical example illustrating the convergence. Last, we generalize our convergence result to a three-timescale RL algorithm introduced by Angiuli et al. to solve mixed Mean Field Control Games (MFCGs).
|
math
|
2,312.04056 |
Safety concerns have risen as robots become more integrated into our daily lives and continue to interact closely with humans. One of the most crucial safety priorities is preventing collisions between robots and people walking nearby. Despite developing various algorithms to address this issue, evaluating their effectiveness on a cost-effective test bench remains a significant challenge. In this work, we propose a solution by introducing a simple yet functional platform that enables researchers and developers to assess how humans interact with mobile robots. This platform is designed to provide a quick yet accurate evaluation of the performance of safe interaction algorithms and make informed decisions for future development. The platform's features and structure are detailed, along with the initial testing results using two preliminary algorithms. The results obtained from the evaluation were consistent with theoretical calculations, demonstrating its effectiveness in assessing human-robot interaction. Our solution provides a preliminary yet reliable approach to ensure the safety of both robots and humans in their daily interactions.
|
eess
|
2,312.05265 |
This paper explores privacy-compliant group-level emotion recognition ''in-the-wild'' within the EmotiW Challenge 2023. Group-level emotion recognition can be useful in many fields including social robotics, conversational agents, e-coaching and learning analytics. This research imposes itself using only global features avoiding individual ones, i.e. all features that can be used to identify or track people in videos (facial landmarks, body poses, audio diarization, etc.). The proposed multimodal model is composed of a video and an audio branches with a cross-attention between modalities. The video branch is based on a fine-tuned ViT architecture. The audio branch extracts Mel-spectrograms and feed them through CNN blocks into a transformer encoder. Our training paradigm includes a generated synthetic dataset to increase the sensitivity of our model on facial expression within the image in a data-driven way. The extensive experiments show the significance of our methodology. Our privacy-compliant proposal performs fairly on the EmotiW challenge, with 79.24% and 75.13% of accuracy respectively on validation and test set for the best models. Noticeably, our findings highlight that it is possible to reach this accuracy level with privacy-compliant features using only 5 frames uniformly distributed on the video.
|
eess
|
2,312.02245 |
This short note is a comment on a historical aspect of a famous formula dating from the 18th century.
|
math
|
2,312.02246 |
Inverse problems aim to determine parameters from observations, a crucial task in engineering and science. Lately, generative models, especially diffusion models, have gained popularity in this area for their ability to produce realistic solutions and their good mathematical properties. Despite their success, an important drawback of diffusion models is their sensitivity to the choice of variance schedule, which controls the dynamics of the diffusion process. Fine-tuning this schedule for specific applications is crucial but time-costly and does not guarantee an optimal result. We propose a novel approach for learning the schedule as part of the training process. Our method supports probabilistic conditioning on data, provides high-quality solutions, and is flexible, proving able to adapt to different applications with minimum overhead. This approach is tested in two unrelated inverse problems: super-resolution microscopy and quantitative phase imaging, yielding comparable or superior results to previous methods and fine-tuned diffusion models. We conclude that fine-tuning the schedule by experimentation should be avoided because it can be learned during training in a stable way that yields better results.
|
cs
|
2,312.0661 |
Many well-studied problems in extremal combinatorics deal with the maximum possible size of a family of objects in which every pair of objects satisfies a given restriction. One problem of this type was recently raised by Alon, Gujgiczer, Körner, Milojević and Simonyi. They asked to determine the maximum size of a family $\mathcal{G}$ of graphs on $[n]$, such that for every two $G_1,G_2 \in \mathcal{G}$, the graphs $G_1 \setminus G_2$ and $G_2 \setminus G_1$ are isomorphic. We completely resolve this problem by showing that this maximum is exactly $2^{\frac{1}{2}\big(\binom{n}{2} - \lfloor \frac{n}{2}\rfloor\big)}$ and characterizing all the extremal constructions. We also prove an analogous result for $r$-uniform hypergraphs.
|
math
|
2,312.05695 |
Vision Transformers have been rapidly uprising in computer vision thanks to their outstanding scaling trends, and gradually replacing convolutional neural networks (CNNs). Recent works on self-supervised learning (SSL) introduce siamese pre-training tasks, on which Transformer backbones continue to demonstrate ever stronger results than CNNs. People come to believe that Transformers or self-attention modules are inherently more suitable than CNNs in the context of SSL. However, it is noteworthy that most if not all prior arts of SSL with CNNs chose the standard ResNets as their backbones, whose architecture effectiveness is known to already lag behind advanced Vision Transformers. Therefore, it remains unclear whether the self-attention operation is crucial for the recent advances in SSL - or CNNs can deliver the same excellence with more advanced designs, too? Can we close the SSL performance gap between Transformers and CNNs? To answer these intriguing questions, we apply self-supervised pre-training to the recently proposed, stronger lager-kernel CNN architecture and conduct an apple-to-apple comparison with Transformers, in their SSL performance. Our results show that we are able to build pure CNN SSL architectures that perform on par with or better than the best SSL-trained Transformers, by just scaling up convolutional kernel sizes besides other small tweaks. Impressively, when transferring to the downstream tasks \texttt{MS COCO} detection and segmentation, our SSL pre-trained CNN model (trained in 100 epochs) achieves the same good performance as the 300-epoch pre-trained Transformer counterpart. We hope this work can help to better understand what is essential (or not) for self-supervised learning backbones.
|
cs
|
2,312.03237 |
Astrochemistry has been widely developed as a power tool to probe physical properties of the interstellar medium (ISM) in various conditions of the Milky Way (MW) Galaxy, and in near and distant galaxies. Most current studies conventionally apply linear scaling to all elemental abundances based on the gas-phase metallicity. However, these elements, including carbon and oxygen, are enriched differentially by stellar nucleosynthesis and the overall galactic chemical evolution, evident from $\alpha$-enhancement in multiple galactic observations such as starbursts, high-redshift star-forming galaxies, and low-metallicity dwarfs. We perform astrochemical modeling to simulate the impact of an $\alpha$-enhanced ISM gas cloud on the abundances of the three phases of carbon (C$^+$, C, CO) dubbed as `the carbon cycle'. The ISM environmental parameters considered include two cosmic-ray ionization rates ($\zeta_{\rm CR}=10^{-17}$ and $10^{-15}\,{\rm s}^{-1}$), two isotropic FUV radiation field strengths ($\chi/\chi_0=1$ and $10^2$), and (sub-)linear dust-to-gas relations against metallicity, mimicking the ISM conditions of different galaxy types. In galaxies with [C/O] $<$ 0, CO, C and C$^+$ all decrease in both abundances and emission, though with differential biases. The low-$J$ CO emission is found to be the most stable tracer for the molecular gas, while C and C$^+$ trace H$_2$ gas only under limited conditions, in line with recent discoveries of [CI]-dark galaxies. We call for caution when using [CII]~$158\mu$m and [CI](1-0) as alternative H$_2$-gas tracers for both diffuse and dense gas with non-zero [C/O] ratios.
|
astro-ph
|
2,312.04403 |
Vision-language pre-training (VLP) models demonstrate impressive abilities in processing both images and text. However, they are vulnerable to multi-modal adversarial examples (AEs). Investigating the generation of high-transferability adversarial examples is crucial for uncovering VLP models' vulnerabilities in practical scenarios. Recent works have indicated that leveraging data augmentation and image-text modal interactions can enhance the transferability of adversarial examples for VLP models significantly. However, they do not consider the optimal alignment problem between dataaugmented image-text pairs. This oversight leads to adversarial examples that are overly tailored to the source model, thus limiting improvements in transferability. In our research, we first explore the interplay between image sets produced through data augmentation and their corresponding text sets. We find that augmented image samples can align optimally with certain texts while exhibiting less relevance to others. Motivated by this, we propose an Optimal Transport-based Adversarial Attack, dubbed OT-Attack. The proposed method formulates the features of image and text sets as two distinct distributions and employs optimal transport theory to determine the most efficient mapping between them. This optimal mapping informs our generation of adversarial examples to effectively counteract the overfitting issues. Extensive experiments across various network architectures and datasets in image-text matching tasks reveal that our OT-Attack outperforms existing state-of-the-art methods in terms of adversarial transferability.
|
cs
|
2,312.03759 |
Large language models (LLMs) are being increasingly incorporated into scientific workflows. However, we have yet to fully grasp the implications of this integration. How should the advent of large language models affect the practice of science? For this opinion piece, we have invited four diverse groups of scientists to reflect on this query, sharing their perspectives and engaging in debate. Schulz et al. make the argument that working with LLMs is not fundamentally different from working with human collaborators, while Bender et al. argue that LLMs are often misused and over-hyped, and that their limitations warrant a focus on more specialized, easily interpretable tools. Marelli et al. emphasize the importance of transparent attribution and responsible use of LLMs. Finally, Botvinick and Gershman advocate that humans should retain responsibility for determining the scientific roadmap. To facilitate the discussion, the four perspectives are complemented with a response from each group. By putting these different perspectives in conversation, we aim to bring attention to important considerations within the academic community regarding the adoption of LLMs and their impact on both current and future scientific practices.
|
cs
|
2,312.05257 |
Autonomous shipping has recently gained much interest in the research community. However, little research focuses on inland - and port navigation, even though this is identified by countries such as Belgium and the Netherlands as an essential step towards a sustainable future. These environments pose unique challenges, since they can contain dynamic obstacles that do not broadcast their location, such as small vessels, kayaks or buoys. Therefore, this research proposes a navigational algorithm which can navigate an inland vessel in a wide variety of complex port scenarios using ranging sensors to observe the environment. The proposed methodology is based on a machine learning approach that has recently set benchmark results in various domains: model-based reinforcement learning. By randomizing the port environments during training, the trained model can navigate in scenarios that it never encountered during training. Furthermore, results show that our approach outperforms the commonly used dynamic window approach and a benchmark model-free reinforcement learning algorithm. This work is therefore a significant step towards vessels that can navigate autonomously in complex port scenarios.
|
cs
|
2,312.03609 |
Resilience is emerging as an evolving notion, reflecting a system's ability to endure and adapt to sudden and catastrophic changes and disruptions. This paper spotlights the significance of the quantitative resilience indices of medium-voltage DC (MVDC) distribution technology in marine vessels, notably naval ships. Given the intricate electrical requirements of modern naval ships, the need for a robust power supply underlines the imperative of resilient DC microgrids. Addressing this, our study introduces a novel quantitative metric for operational resilience of DC microgrids based on the measured voltage of main DC bus. This metric not only fuses real-time tracking, compatibility, and computational efficiency, but also adeptly monitors multiple event phases based on time-domain analysis of dc bus voltage dynamics. The intricacies of the dc bus voltage, including overshoots and undershoots, are meticulously accounted for in the algorithm design. With respect to existing research that typically focuses on offline resilience assessments, the proposed index provides valuable real-time information for microgrid operators and identifies whether microgrid resilience is deteriorating over time.
|
eess
|
2,312.02873 |
The process engineering domain widely uses Process Flow Diagrams (PFDs) and Process and Instrumentation Diagrams (P&IDs) to represent process flows and equipment configurations. However, the P&IDs and PFDs, hereafter called flowsheets, can contain errors causing safety hazards, inefficient operation, and unnecessary expenses. Correcting and verifying flowsheets is a tedious, manual process. We propose a novel generative AI methodology for automatically identifying errors in flowsheets and suggesting corrections to the user, i.e., autocorrecting flowsheets. Inspired by the breakthrough of Large Language Models (LLMs) for grammatical autocorrection of human language, we investigate LLMs for the autocorrection of flowsheets. The input to the model is a potentially erroneous flowsheet and the output of the model are suggestions for a corrected flowsheet. We train our autocorrection model on a synthetic dataset in a supervised manner. The model achieves a top-1 accuracy of 80% and a top-5 accuracy of 84% on an independent test dataset of synthetically generated flowsheets. The results suggest that the model can learn to autocorrect the synthetic flowsheets. We envision that flowsheet autocorrection will become a useful tool for chemical engineers.
|
cs
|
2,312.05184 |
We explore the capability of measuring lensing signals in $LiteBIRD$ full-sky polarization maps. With a $30$ arcmin beam width and an impressively low polarization noise of $2.16\,\mu$K-arcmin, $LiteBIRD$ will be able to measure the full-sky polarization of the cosmic microwave background (CMB) very precisely. This unique sensitivity also enables the reconstruction of a nearly full-sky lensing map using only polarization data, even considering its limited capability to capture small-scale CMB anisotropies. In this paper, we investigate the ability to construct a full-sky lensing measurement in the presence of Galactic foregrounds, finding that several possible biases from Galactic foregrounds should be negligible after component separation by harmonic-space internal linear combination. We find that the signal-to-noise ratio of the lensing is approximately $40$ using only polarization data measured over $90\%$ of the sky. This achievement is comparable to $Planck$'s recent lensing measurement with both temperature and polarization and represents a four-fold improvement over $Planck$'s polarization-only lensing measurement. The $LiteBIRD$ lensing map will complement the $Planck$ lensing map and provide several opportunities for cross-correlation science, especially in the northern hemisphere.
|
astro-ph
|
2,312.05977 |
This paper axiomatizes, in a two-stage setup, a new theory for decision under risk and ambiguity. The axiomatized preference relation $\succeq$ on the space $\tilde{V}$ of random variables induces an ambiguity index $c$ on the space $\Delta$ of probabilities, a probability weighting function $\psi$, generating the measure $\nu_{\psi}$ by transforming an objective probability measure, and a utility function $\phi$, such that, for all $\tilde{v},\tilde{u}\in\tilde{V}$, \begin{align*} \tilde{v}\succeq\tilde{u} \Leftrightarrow \min_{Q \in \Delta} \left\{\mathbb{E}_Q\left[\int\phi\left(\tilde{v}^{\centerdot}\right)\,\mathrm{d}\nu_{\psi}\right]+c(Q)\right\} \geq \min_{Q \in \Delta} \left\{\mathbb{E}_Q\left[\int\phi\left(\tilde{u}^{\centerdot}\right)\,\mathrm{d}\nu_{\psi}\right]+c(Q)\right\}. \end{align*} Our theory extends the rank-dependent utility model of Quiggin (1982) for decision under risk to risk and ambiguity, reduces to the variational preferences model when $\psi$ is the identity, and is dual to variational preferences when $\phi$ is affine in the same way as the theory of Yaari (1987) is dual to expected utility. As a special case, we obtain a preference axiomatization of a decision theory that is a rank-dependent generalization of the popular maxmin expected utility theory. We characterize ambiguity aversion in our theory.
|
math
|
2,312.0656 |
In this work, we consider the problem of regularization in minimum mean-squared error (MMSE) linear filters. Exploiting the relationship with statistical machine learning methods, the regularization parameter is found from the observed signals in a simple and automatic manner. The proposed approach is illustrated through system identification examples, where the automatic regularization yields near-optimal results.
|
cs
|
2,312.03262 |
We present a robust membership inference attack (RMIA) that amplifies the distinction between population data and the training data on any target model, by effectively leveraging both reference models and reference data in our likelihood ratio test. Our algorithm exhibits superior test power (true-positive rate) when compared to prior methods, even at extremely low false-positive error rates (as low as 0). Also, under computation constraints, where only a limited number of reference models (as few as 1) are available, our method performs exceptionally well, unlike some prior attacks that approach random guessing in such scenarios. Our method lays the groundwork for cost-effective and practical yet powerful and robust privacy risk analysis of machine learning algorithms.
|
cs
|
2,312.03166 |
Inferring parameters of macro-kinetic growth models, typically represented by Ordinary Differential Equations (ODE), from the experimental data is a crucial step in bioprocess engineering. Conventionally, estimates of the parameters are obtained by fitting the mechanistic model to observations. Fitting, however, requires a significant computational power. Specifically, during the development of new bioprocesses that use previously unknown organisms or strains, efficient, robust, and computationally cheap methods for parameter estimation are of great value. In this work, we propose using Deep Neural Networks (NN) for directly predicting parameters of mechanistic models given observations. The approach requires spending computational resources for training a NN, nonetheless, once trained, such a network can provide parameter estimates orders of magnitude faster than conventional methods. We consider a training procedure that combines Neural Networks and mechanistic models. We demonstrate the performance of the proposed algorithms on data sampled from several mechanistic models used in bioengineering describing a typical industrial batch process and compare the proposed method, a typical gradient-based fitting procedure, and the combination of the two. We find that, while Neural Network estimates are slightly improved by further fitting, these estimates are measurably better than the fitting procedure alone.
|
cs
|
2,312.04014 |
Green hydrogen has shown great potential to power microgrids as a primary source, yet the operation methodology under extreme events is still an open area. To fill this gap, this letter establishes an operational optimization strategy towards resilient hydrogen-powered microgrids, where the frequency and voltage regulation characteristics of hydrogen sources under advanced controls are accurately represented by piecewise linear constraints. The results show that the new operation approach can output a safety-assured operation plan with rational power change distribution and reduced frequency and voltage variation.
|
eess
|
2,312.02487 |
The DOA estimation method of coherent signals based on periodical coding metasurface is proposed. After periodical coding, the DOA information of incident signals in the time domain is represented as the amplitude and phase information at different frequency points in the frequency domain. Finite time Fourier transform (FTFT) is performed on the received signal and appropriate frequency points are selected to reconstruct the frequency domain snapshot, then pattern smoothing (PS) technique is applied to execute DOA estimation. Compared with conventional DOA estimation methods, the proposed method has two main advantages: one is that only a single receiving channel is needed to avoid the appearance of channel mismatch errors, the other is that it can process with multiple coherent signals. The performance curves of the proposed method are analyzed under different conditions and compared with existing methods. Simulation results show the effectiveness of the proposed method.
|
eess
|
2,312.03057 |
An overarching milestone of quantum machine learning (QML) is to demonstrate the advantage of QML over all possible classical learning methods in accelerating a common type of learning task as represented by supervised learning with classical data. However, the provable advantages of QML in supervised learning have been known so far only for the learning tasks designed for using the advantage of specific quantum algorithms, i.e., Shor's algorithms. Here we explicitly construct an unprecedentedly broader family of supervised learning tasks with classical data to offer the provable advantage of QML based on general quantum computational advantages, progressing beyond Shor's algorithms. Our learning task is feasibly achievable by executing a general class of functions that can be computed efficiently in polynomial time for a large fraction of inputs by arbitrary quantum algorithms but not by any classical algorithm. We prove the hardness of achieving this learning task for any possible polynomial-time classical learning method. We also clarify protocols for preparing the classical data to demonstrate this learning task in experiments. These results open routes to exploit a variety of quantum advantages in computing functions for the experimental demonstration of the advantage of QML.
|
cs
|
2,312.03878 |
Machine learning models are often trained to predict the outcome resulting from a human decision. For example, if a doctor decides to test a patient for disease, will the patient test positive? A challenge is that the human decision censors the outcome data: we only observe test outcomes for patients doctors historically tested. Untested patients, for whom outcomes are unobserved, may differ from tested patients along observed and unobserved dimensions. We propose a Bayesian model class which captures this setting. The purpose of the model is to accurately estimate risk for both tested and untested patients. Estimating this model is challenging due to the wide range of possibilities for untested patients. To address this, we propose two domain constraints which are plausible in health settings: a prevalence constraint, where the overall disease prevalence is known, and an expertise constraint, where the human decision-maker deviates from purely risk-based decision-making only along a constrained feature set. We show theoretically and on synthetic data that domain constraints improve parameter inference. We apply our model to a case study of cancer risk prediction, showing that the model's inferred risk predicts cancer diagnoses, its inferred testing policy captures known public health policies, and it can identify suboptimalities in test allocation. Though our case study is in healthcare, our analysis reveals a general class of domain constraints which can improve model estimation in many settings.
|
cs
|
2,312.04017 |
The discovery of gravitational waves (GWs) opens a new window for exploring the physics of the early universe. Identifying the source of GWs and their spectra today turn out to be the important tasks so as to assist the experimental detection of stochastic GWs. In this paper, we investigate the oscillations of the ultralight dark photon (ULDP) into GWs in the dark halo. Assuming dark matter is composed of the ULDP and there are primordial dark magnetic fields (PDMFs) arising from the axion inflation and/or the dark phase transition, then the ULDP can oscillate into the GW when it passes through an environment of PDMFs. We derive the local energy density of GWs in the galaxy cluster induced by the instaneous oscillation of ULDP in the PDMFs. These stochastic local GWs exhibit a pulse-like spectrum, with frequency depending on the mass of the ULDP, and can be detected in Pulsar Timing Arrays (PTAs) or future space-based interferometers. We also find that the low-frequency GW signal observed by the NANOGrav collaboration and other PTA experiments can be explained by the oscillation of the ULDP in the PDMFs in the early universe.
|
astro-ph
|
2,312.05516 |
Large Language Models (LLMs) have recently experienced great success, as evident in the widespread popularity of ChatGPT. Existing LLM serving systems are stateless across requests. Consequently, when LLMs are used in the common setting of multi-turn conversations, a growing log of the conversation history must be processed alongside any request by the serving system at each turn, resulting in repeated history processing. In this paper, we design $Pensieve$, a system optimized for multi-turn conversation LLM serving. $Pensieve$ maintains the conversation state across requests by caching previously processed history to avoid duplicate processing. $Pensieve$'s multi-tier caching strategy can utilize both GPU and CPU memory to efficiently store and retrieve cached data. $Pensieve$ also generalizes the recent PagedAttention kernel to support attention between multiple input tokens with a GPU cache spread over non-contiguous memory. Our evaluation shows that $Pensieve$ is able to achieve 1.51-1.95x throughput compared to vLLM and reduce latency by 60-75%.
|
cs
|
2,312.05452 |
Matter-wave interferometers with micro-particles are excellent quantum sensors as they can be sensitive to a minute quantum phase information, which a classical detector cannot. Two such adjacent micro-particles in the interferometers can be entangled solely via the quantum nature of gravity known as the quantum gravity-induced entanglement of masses (QGEM) protocol. The micro-particles can also be entangled via EM interactions. Therefore, it is essential to estimate the decoherence, noise/dephasing rate for such interferometers. In this paper, we will focus on a particular source of an electromagnetic interaction. We will treat this interaction as a noise which will lead to dephasing. We assume that our matter-wave interferometer has a residual charge which can interact with a neighbouring ion in the ambience, e.g., inside the experimental capsule. This will provide dephasing of the matter-wave interferometer due to the Coulomb interaction with external charges and the charge-dipole interaction with external dielectrics or dipoles. Similarly, we will consider neutral micro-particles, which can interact with charged and/or neutral particles in the ambience via induced dipole-charge, permanent dipole-charge, and dipole-dipole interactions. All these interactions constitute electromagnetically driven dephasing to a single and a twin interferometer. We will discuss their relevance for the QGEM experiment and provide insight into the noise of an entangled state for charged micro-particles kept adjacently with an implication for the C-NOT gate.
|
quant-ph
|
2,312.0659 |
General relativity's prediction that all black holes share identical properties, irrespective of their size, can now be empirically tested using electromagnetic observations of supermassive black holes and gravitational waves from mergers of stellar-mass black holes. In this work, we focus on the electromagnetic side of this test and quantify the constraining power of very-long-baseline interferometry (VLBI). We demonstrate how to use lensing bands -- annular regions on the observer's screen surrounding the critical curve -- to constrain the underlying spacetime geometry. Contingent upon a detection of a lensed VLBI feature, the resulting lensing-band framework allows us to exclude spacetimes for which said feature cannot arise from geodesics that traversed the equatorial plane more than once. Focusing on the first lensed image and tests of black-hole uniqueness, we employ a parametrized spacetime as a case study. We find that resolving geometric information that goes beyond the apparent size of the critical curve has the potential to lift degeneracies between different spacetime parameters. Our work thereby quantifies a conservative estimate of the constraining power of VLBI measurements and contributes to a larger effort to simultaneously constrain geometry and astrophysics.
|
astro-ph
|
2,312.02807 |
We develop the information geometry of scaled Gaussian distributions for which the covariance matrix exhibits a Kronecker product structure. This model and its geometry are then used to propose an online change detection (CD) algorithm for multivariate image times series (MITS). The proposed approach relies mainly on the online estimation of the structured covariance matrix under the null hypothesis, which is performed through a recursive (natural) Riemannian gradient descent. This approach exhibits a practical interest compared to the corresponding offline version, as its computational cost remains constant for each new image added in the time series. Simulations show that the proposed recursive estimators reach the Intrinsic Cramér-Rao bound. The interest of the proposed online CD approach is demonstrated on both simulated and real data.
|
stat
|
2,312.05027 |
The presence of $Gaia$ DR3 provides a large sample of stars with complete 6D information, offering a fertile ground for the exploration of stellar objects that were accreted to the Milky Way through ancient merger events. In this study, we developed a deep learning methodology to identify ex-situ stars within the $Gaia$ DR3 catalogue. After two phases of training, our neural network (NN) model was capable of performing binary classification of stars based on input data consisting of 3D position and velocity, as well as actions. From the target sample of 27 085 748 stars, our NN model managed to identify 160 146 ex-situ stars. The metallicity distribution suggests that this ex-situ sample comprises multiple components but appears to be predominated by the Gaia-Sausage-Enceladus. We identified member stars of the Magellanic Clouds, Sagittarius, and 20 globular clusters throughout our examination. Furthermore, an extensive group of member stars from Gaia-Sausage-Enceladus, Thamnos, Sequoia, Helmi streams, Wukong, and Pontus were meticulously selected, constituting an ideal sample for the comprehensive study of substructures. Finally, we conducted a preliminary estimation to determine the proportions of ex-situ stars in the thin disc, thick disc, and halo, which resulted in percentages of 0.1%, 1.6%, and 63.2%, respectively. As the vertical height from the Galactic disc and distance from the Galactic centre increased, there was a corresponding upward trend in the ex-situ fraction of the target sample.
|
astro-ph
|
2,312.06132 |
We give explicit expressions for MacMahon's generalized sums-of-divisors $q$-series $A_r$ and $C_r$ by relating them to (odd) multiple Eisenstein series. Recently, these sums-of-divisors have been studied in the context of quasimodular forms, vertex algebras, $N=4$ $SU(N)$ Super-Yang-Mills theory, and the study of congruences of partitions. We relate them to a broader mathematical framework and give explicit expressions for both $q$-series in terms of Eisenstein series and their odd variants.
|
math
|
2,312.06238 |
In [6], Geroch, Kronheimer and Penrose introduced a way to attach ideal points to a spacetime M , defining the causal completion of M. They established that this is a topological space which is Hausdorff when M is globally hyperbolic. In this paper, we prove that if, in addition, M is simply-connected and conformally flat, its causal completion is a topological manifold with boundary homeomorphic to S x [0, 1] where S is a Cauchy hypersurface of M. We also introduce three remarkable families of globally hyperbolic conformally flat spacetimes and provide a description of their causal completions.
|
math
|
2,312.0598 |
With the increasing effects of climate change, the urgency to step away from fossil fuels is greater than ever before. Electric vehicles (EVs) are one way to diminish these effects, but their widespread adoption is often limited by the insufficient availability of charging stations. In this work, our goal is to expand the infrastructure of EV charging stations, in order to provide a better quality of service in terms of user satisfaction (and availability of charging stations). Specifically, our focus is directed towards urban areas. We first propose a model for the assignment of EV charging demand to stations, framing it as a maximum flow problem. This model is the basis for the evaluation of user satisfaction with a given charging infrastructure. Secondly, we incorporate the maximum flow model into a mixed-integer linear program, where decisions on the opening of new stations and on the expansion of their capacity through additional outlets is accounted for. We showcase our methodology for the city of Montreal, demonstrating the scalability of our approach to handle real-world scenarios. We conclude that considering both spacial and temporal variations in charging demand is meaningful when solving realistic instances.
|
cs
|
2,312.0269 |
Autonomous underwater vehicles (AUV) have become the de facto vehicle for remote operations involving oceanography, inspection, and monitoring tasks. These vehicles operate in different and often challenging environments; hence, the design and development of the AUV involving hydrodynamics and control systems need to be designed in detail. This book chapter presents a study on the modelling and robust control of a research vehicle in the presence of uncertainties. The vehicle's dynamic behaviour is modelled using a 6-degree-of-freedom approach, considering the effect of ocean currents. The level flight requirements for different speeds are derived, and the resulting model is decomposed into horizontal and vertical subsystems for linear analysis. The simulation results presented focus on the efficacy of linear controllers within three key subsystems: depth, yaw, and speed. Moreover, level-flight outcomes are demonstrated for a speed of 4 knots. The nonlinear control strategies employed in this study encompass conventional and sliding-mode control (SMC) methodologies. To ensure accurate tracking performance, the controller design considers the vehicle's dynamics with various uncertainties such as ocean currents, parameter uncertainty, CG (Center of Gravity) deviation and buoyancy variation. Both conventional and nonlinear SMC controllers' outcomes are showcased with a lawn-mowing manoeuvre scenario. A systematic comparison is drawn between the robustness of SMC against disturbances and parameter fluctuations in contrast to conventional controllers. Importantly, these results underscore the trade-off that accompanies SMC's robustness, as it necessitates a higher level of complexity in terms of controller design, intricate implementation intricacies, and the management of chattering phenomena.
|
eess
|
2,312.03817 |
We explore the problem of computationally generating special `prime' images that produce optical illusions when physically arranged and viewed in a certain way. First, we propose a formal definition for this problem. Next, we introduce Diffusion Illusions, the first comprehensive pipeline designed to automatically generate a wide range of these illusions. Specifically, we both adapt the existing `score distillation loss' and propose a new `dream target loss' to optimize a group of differentially parametrized prime images, using a frozen text-to-image diffusion model. We study three types of illusions, each where the prime images are arranged in different ways and optimized using the aforementioned losses such that images derived from them align with user-chosen text prompts or images. We conduct comprehensive experiments on these illusions and verify the effectiveness of our proposed method qualitatively and quantitatively. Additionally, we showcase the successful physical fabrication of our illusions -- as they are all designed to work in the real world. Our code and examples are publicly available at our interactive project website:
|
cs
|
2,312.02711 |
For safe and effective operation of humanoid robots in human-populated environments, the problem of commanding a large number of Degrees of Freedom (DoF) while simultaneously considering dynamic obstacles and human proximity has still not been solved. We present a new reactive motion controller that commands two arms of a humanoid robot and three torso joints (17 DoF in total). We formulate a quadratic program that seeks joint velocity commands respecting multiple constraints while minimizing the magnitude of the velocities. We introduce a new unified treatment of obstacles that dynamically maps visual and proximity (pre-collision) and tactile (post-collision) obstacles as additional constraints to the motion controller, in a distributed fashion over surface of the upper-body of the iCub robot (with 2000 pressure-sensitive receptors). The bio-inspired controller: (i) produces human-like minimum jerk movement profiles; (ii) gives rise to a robot with whole-body visuo-tactile awareness, resembling peripersonal space representations. The controller was extensively experimentally validated, including a physical human-robot interaction scenario.
|
cs
|
2,312.02778 |
Reentrant localization (RL), a recently prominent phenomenon, traditionally links to the interplay of staggered correlated disorder and hopping dimerization, as indicated by prior research. Contrary to this paradigm, our present study demonstrates that hopping dimerization is not a pivotal factor in realizing RL. Considering a helical magnetic system with antiferromagnetic ordering, we uncover spin-dependent RL at multiple energy regions, in the {\em absence} of hopping dimerization. This phenomenon persists even in the thermodynamic limit. The correlated disorder in the form of Aubry-André-Harper model is introduced by applying a transverse electric field to the helical system, circumventing the use of traditional substitutional disorder. Described within a tight-binding framework, present work provides a novel outlook on RL, highlighting the crucial role of electric field, antiferromagnetic ordering, and the helicity of the geometry.
|
quant-ph
|
2,312.05034 |
Grasp force synthesis is a non-convex optimization problem involving constraints that are bilinear. Traditional approaches to this problem involve general-purpose gradient-based nonlinear optimization and semi-definite programming. With a view towards dealing with postural synergies and non-smooth but convex positive semidefinite constraints, we look beyond gradient-based optimization. The focus of this paper is to undertake a grasp analysis of biomimetic grasping in multi-fingered robotic hands as a bilinear matrix inequality (BMI) problem. Our analysis is to solve it using a deep learning approach to make the algorithm efficiently generate force closure grasps with optimal grasp quality on untrained/unseen objects.
|
cs
|
2,312.02283 |
Recent observations of galaxy clusters and groups with misalignments between their central AGN jets and X-ray cavities, or with multiple misaligned cavities, have raised concerns about the jet - bubble connection in cooling cores, and the processes responsible for jet realignment. To investigate the frequency and causes of such misalignments, we construct a sample of 16 cool core galaxy clusters and groups. Using VLBA radio data we measure the parsec-scale position angle of the jets, and compare it with the position angle of the X-ray cavities detected in Chandra data. Using the overall sample and selected subsets, we consistently find that there is a 30% - 38% chance to find a misalignment larger than $\Delta\Psi = 45^{\circ}$ when observing a cluster/group with a detected jet and at least one cavity. We determine that projection may account for an apparently large $\Delta\Psi$ only in a fraction of objects ($\sim$35%), and given that gas dynamical disturbances (as sloshing) are found in both aligned and misaligned systems, we exclude environmental perturbation as the main driver of cavity - jet misalignment. Moreover, we find that large misalignments (up to $\sim90^{\circ}$) are favored over smaller ones ($45^{\circ}\leq\Delta\Psi\leq70^{\circ}$), and that the change in jet direction can occur on timescales between one and a few tens of Myr. We conclude that misalignments are more likely related to actual reorientation of the jet axis, and we discuss several engine-based mechanisms that may cause these dramatic changes.
|
astro-ph
|
2,312.05176 |
We consider a missing data problem in the context of automatic segmentation methods for Magnetic Resonance Imaging (MRI) brain scans. Usually, automated MRI scan segmentation is based on multiple scans (e.g., T1-weighted, T2-weighted, T1CE, FLAIR). However, quite often a scan is blurry, missing or otherwise unusable. We investigate the question whether a missing scan can be synthesized. We exemplify that this is in principle possible by synthesizing a T2-weighted scan from a given T1-weighted scan. Our first aim is to compute a picture that resembles the missing scan closely, measured by average mean squared error (MSE). We develop/use several methods for this, including a random baseline approach, a clustering-based method and pixel-to-pixel translation method by (Pix2Pix) which is based on conditional GANs. The lowest MSE is achieved by our clustering-based method. Our second aim is to compare the methods with respect to the affect that using the synthesized scan has on the segmentation process. For this, we use a DeepMedic model trained with the four input scan modalities named above. We replace the T2-weighted scan by the synthesized picture and evaluate the segmentations with respect to the tumor identification, using Dice scores as numerical evaluation. The evaluation shows that the segmentation works well with synthesized scans (in particular, with Pix2Pix methods) in many cases.
|
cs
|
2,312.06236 |
Investors are interested in predicting future success of startup companies, preferably using publicly available data which can be gathered using free online sources. Using public-only data has been shown to work, but there is still much room for improvement. Two of the best performing prediction experiments use 17 and 49 features respectively, mostly numeric and categorical in nature. In this paper, we significantly expand and diversify both the sources and the number of features (to 171) to achieve better prediction. Data collected from Crunchbase, the Google Search API, and Twitter (now X) are used to predict whether a company will raise a round of funding within a fixed time horizon. Much of the new features are textual and the Twitter subset include linguistic metrics such as measures of passive voice and parts-of-speech. A total of ten machine learning models are also evaluated for best performance. The adaptable model can be used to predict funding 1-5 years into the future, with a variable cutoff threshold to favor either precision or recall. Prediction with comparable assumptions generally achieves F scores above 0.730 which outperforms previous attempts in the literature (0.531), and does so with fewer examples. Furthermore, we find that the vast majority of the performance impact comes from the top 18 of 171 features which are mostly generic company observations, including the best performing individual feature which is the free-form text description of the company.
|
cs
|
2,312.03885 |
We consider a gradient-based optimization method applied to a function $\mathcal{L}$ of a vector of variables $\boldsymbol{\theta}$, in the case where $\boldsymbol{\theta}$ is represented as a tuple of tensors $(\mathbf{T}_1, \cdots, \mathbf{T}_S)$. This framework encompasses many common use-cases, such as training neural networks by gradient descent. First, we propose a computationally inexpensive technique providing higher-order information on $\mathcal{L}$, especially about the interactions between the tensors $\mathbf{T}_s$, based on automatic differentiation and computational tricks. Second, we use this technique at order 2 to build a second-order optimization method which is suitable, among other things, for training deep neural networks of various architectures. This second-order method leverages the partition structure of $\boldsymbol{\theta}$ into tensors $(\mathbf{T}_1, \cdots, \mathbf{T}_S)$, in such a way that it requires neither the computation of the Hessian of $\mathcal{L}$ according to $\boldsymbol{\theta}$, nor any approximation of it. The key part consists in computing a smaller matrix interpretable as a "Hessian according to the partition", which can be computed exactly and efficiently. In contrast to many existing practical second-order methods used in neural networks, which perform a diagonal or block-diagonal approximation of the Hessian or its inverse, the method we propose does not neglect interactions between layers. Finally, we can tune the coarseness of the partition to recover well-known optimization methods: the coarsest case corresponds to Cauchy's steepest descent method, the finest case corresponds to the usual Newton's method.
|
cs
|
2,312.0319 |
In two parts, we present a bigness criterion for the cotangent bundle of resolutions of orbifold surfaces of general type. As a corollary, we obtain the \textit{canonical model singularities} (CMS) criterion that can be applied to determine when a birational class of surfaces has smooth representatives with big cotangent bundle and compare it with other known criteria. We then apply this criterion to the problem of finding the minimal degrees $d$ for which the deformation equivalence class of a smooth hypersurface of degree d in $\PP^3$ has a representative with big cotangent bundle; applications to the minimal resolutions of cyclic covers of $\PP^2$ branched along line arrangements in general position are also obtained.
The CMS criterion involves invariants of canonical singularities whose values were unknown. In this part of the work, we describe a method of finding these invariants and obtain formulas for $A_n$-singularities. We also use our approach to derive several extension results for symmetric differentials on the complement of the exceptional locus $E$ of the minimal resolution of an $A_n$-singularity; namely, we characterize the precise extent to which the poles along $E$ of symmetric differentials on the complement are milder than logarithmic poles.
|
math
|
2,312.04329 |
This paper shows that a class of codes such as Reed-Muller (RM) codes have vanishing bit-error probability below capacity on symmetric channels. The proof relies on the notion of `camellia codes': a class of symmetric codes decomposable into `camellias', i.e., set systems that differ from sunflowers by allowing for scattered petal overlaps. The proof then follows from a boosting argument on the camellia petals with second moment Fourier analysis. For erasure channels, this gives a self-contained proof of the bit-error result in Kudekar et al.'17, without relying on sharp thresholds for monotone properties Friedgut-Kalai'96. For error channels, this gives a shortened proof of Reeves-Pfister'23 with an exponentially tighter bound, and a proof variant of the bit-error result in Abbe-Sandon'23. The control of the full (block) error probability still requires Abbe-Sandon'23 for RM codes.
|
cs
|
2,312.04039 |
We only consider finite structures. With every totally ordered set $V$ and a subset $P$ of $\binom{V}{2}$, we associate the underlying tournament ${\rm Inv}(\underline{V}, P)$ obtained from the transitive tournament $\underline{V}:=(V, \{(x,y) \in V \times V : x < y \})$ by reversing $P$, i.e., by reversing the arcs $(x,y)$ such that $\{x,y\} \in P$. The subset $P$ is a pairing (of $\cup P$) if $|\cup P| = 2|P|$, a quasi-pairing (of $\cup P$) if $|\cup P| = 2|P|-1$; it is irreducible if no nontrivial interval of $\cup P$ is a union of connected components of the graph $(\cup P, P)$. In this paper, we consider pairings and quasi-pairings in relation to tournaments. We establish close relationships between irreducibility of pairings (or quasi-pairings) and indecomposability of their underlying tournaments under modular decomposition. For example, given a pairing $P$ of a totally ordered set $V$ of size at least $6$, the pairing $P$ is irreducible if and only if the tournament ${\rm Inv}(\underline{V}, P)$ is indecomposable. This is a consequence of a more general result characterizing indecomposable tournaments obtained from transitive tournaments by reversing pairings. We obtain analogous results in the case of quasi-pairings.
|
math
|
2,312.04523 |
Branched rough paths define integration theories that may fail to satisfy the usual integration by parts identity. The intrinsically-defined projection of the Connes-Kreimer Hopf algebra onto its primitive elements defined by Broadhurst and Kreimer, and further studied by Foissy, allows us to view it as a commutative $\mathbf{B}_\infty$-algebra and thus to write an explicit change-of-variable formula for solutions to rough differential equations. This formula, which is realised by means of an explicit morphism from the Grossman-Larson Hopf algebra to the Hopf algebra of differential operators, restricts to the well-known Itô formula for semimartingales. We stablish an isomorphism with the shuffle algebra over primitives, extending Hoffman's exponential for the quasi shuffle algebra, and in particular the usual Itô-Stratonovich correction formula for semimartingales. We place special emphasis on the one-dimensional case, in which certain rough path terms can be expressed as polynomials in the extended trace path, which for semimartingales restrict to the well-known Kailath-Segall polynomials. We end by describing an algebraic framework for generating examples of branched rough paths, and, motivated by the recent literature on stochastic processes, exhibit a few such examples above scalar 1/4-fractional Brownian motion, two of which are "truly branched", i.e. not quasi-geometric.
|
math
|
2,312.03984 |
Many polyatomic astrophysical plasmas are compressible and out of chemical and thermal equilibrium, and yet, due to Stokes' hypothesis, a means to carefully control the decay of compressible modes in these systems has largely been neglected. This is especially important for small-scale, turbulent dynamo processes, which are known to be sensitive to the effects of compression. To control the viscous properties of the compressible modes, we perform supersonic, visco-resistive dynamo simulations with additional bulk viscosity $\nu_{\rm bulk}$, deriving a new $\nu_{\rm bulk}$ Reynolds number $\rm{Re}_{\rm bulk}$, and viscous Prandtl number $\rm{P}\nu \equiv \rm{Re}_{\rm bulk} / \rm{Re}_{\rm shear}$, where $\rm{Re}_{\rm shear}$ is the shear viscosity Reynolds number. For $10^{-3} \leq \rm{P}\nu \leq \infty$, we explore a broad range of statistics critical to the dynamo problem, including the integral and spectral energy ratios, growth rates, and the magnetic $E_{\rm mag}(k)$ and kinetic $E_{\rm kin}(k)$ energy spectrum. We derive a general framework for decomposing $E_{\rm mag}$ growth rates into incompressible and compressible terms via orthogonal tensor decompositions of $\nabla\otimes\mathbf{v}$, where $\mathbf{v}$ is the fluid velocity. We find that compressible modes play a dual role, growing and decaying $E_{\rm mag}$, and that field-line stretching is the main driver of growth, even in supersonic dynamos. In the absence of $\nu_{\rm bulk}$, compressible modes pile up on small-scales, creating an apparent spectral bottleneck, which disappears for $\rm{P}\nu \approx 1$. As $\rm{P}\nu$ decreases, compressible modes are dissipated at increasingly larger scales, in turn suppressing incompressible modes through a coupling between viscosity operators. We emphasise the importance of further understanding the role of $\nu_{\rm bulk}$ in compressible astrophysical plasmas.
|
astro-ph
|
2,312.06481 |
Socially Assistive Robots represent a valid solution for improving the quality of life and the mood of older adults. In this context, this work presents the CloudIA robot, a non-human-like robot intended to promote sociality and well-being among older adults. The design of the robot and of the provided services were carried out by a multidisciplinary team of designers and technology developers in tandem with professional caregivers. The capabilities of the robot were implemented according to the received guidelines and tested in two nursing facilities by 15 older people. Qualitative and quantitative metrics were used to investigate the engagement of the participants during the interaction with the robot, and to investigate any differences in the interaction during the proposed activities. The results highlighted the general tendency of humanizing the robotic platform and demonstrated the feasibility of introducing the CloudIA robot in support of the professional caregivers' work. From this pilot test, further ideas on improving the personalization of the robotic platform emerged.
|
cs
|
2,312.04355 |
This paper studies a secure cell-free integrated sensing and communication (ISAC) system, in which multiple ISAC transmitters collaboratively send confidential information to multiple communication users (CUs) and concurrently conduct target detection. Different from prior works investigating communication security against potential information eavesdropping, we consider the security of both communication and sensing in the presence of both information and sensing eavesdroppers that aim to intercept confidential communication information and extract target information, respectively. Towards this end, we optimize the joint information and sensing transmit beamforming at these ISAC transmitters for secure cell-free ISAC. Our objective is to maximize the detection probability over a designated sensing area while ensuring the minimum signal-to-interference-plus-noise-ratio (SINR) requirements at CUs. Our formulation also takes into account the maximum tolerable signal-to-noise ratio (SNR) at information eavesdroppers for ensuring the confidentiality of information transmission, and the maximum detection probability constraints at sensing eavesdroppers for preserving sensing privacy. The formulated secure joint transmit beamforming problem is highly non-convex due to the intricate interplay between the detection probabilities, beamforming vectors, and SINR constraints. Fortunately, through strategic manipulation and via applying the semidefinite relaxation (SDR) technique, we successfully obtain the globally optimal solution to the design problem by rigorously verifying the tightness of SDR. Furthermore, we present two alternative joint beamforming designs based on the sensing SNR maximization over the specific sensing area and the coordinated beamforming, respectively. Numerical results reveal the benefits of our proposed design over these alternative benchmarks.
|
math
|
2,312.04566 |
Recently diffusion models have shown improvement in synthetic image quality as well as better control in generation. We motivate and present Gen2Det, a simple modular pipeline to create synthetic training data for object detection for free by leveraging state-of-the-art grounded image generation methods. Unlike existing works which generate individual object instances, require identifying foreground followed by pasting on other images, we simplify to directly generating scene-centric images. In addition to the synthetic data, Gen2Det also proposes a suite of techniques to best utilize the generated data, including image-level filtering, instance-level filtering, and better training recipe to account for imperfections in the generation. Using Gen2Det, we show healthy improvements on object detection and segmentation tasks under various settings and agnostic to detection methods. In the long-tailed detection setting on LVIS, Gen2Det improves the performance on rare categories by a large margin while also significantly improving the performance on other categories, e.g. we see an improvement of 2.13 Box AP and 1.84 Mask AP over just training on real data on LVIS with Mask R-CNN. In the low-data regime setting on COCO, Gen2Det consistently improves both Box and Mask AP by 2.27 and 1.85 points. In the most general detection setting, Gen2Det still demonstrates robust performance gains, e.g. it improves the Box and Mask AP on COCO by 0.45 and 0.32 points.
|
cs
|
2,312.03779 |
Given the widespread popularity of interactive AI models like ChatGPT, public opinion on emerging artificial intelligence generated content(AIGC) has been extensively debated. Pessimists believe that AIGC will replace humans in the future, and optimists think that it will further liberate productivity. Public emotions play a crucial role on social media platforms. They can provide valuable insights into the public's opinions, attitudes, and behaviors. There is a lack of research on the analysis of social group emotions triggered by AIGC content, and even more on the cross-platform differences of group emotions. This study fills the research gap by connecting the theory of group dynamics with emotions in social media. Specifically, we develop a scientific group emotion calculation and visualization system based on chains of communication. The system is capable of crawling data in real time and presenting the current state of group emotions in a fine-grained manner. We then analyze which group dynamic factors drive different public emotions towards nine AIGC products on the three most popular social media platforms in China. Finally, we obtain four main findings. First, Douyin is the only platform with negative group emotion on emerging AI technologies. Second, Weibo users prefer extreme emotions more than others. Third, the group emotion varies by education and age. It is negatively correlated with senior high school or lower and 25 or younger, and positively correlated with bachelor's degree or higher and 26-35. Fourth, the group emotion polarization increases with more posts without comments and celebrity publishers. By analyzing the key dynamic factors of group emotions to AIGC on various social media platforms, we can improve our products and services, develop more effective marketing strategies, and create more accurate and effective AI models to solve complex problems.
|
cs
|
2,312.05942 |
We investigate how well a simple leading order perturbation theory model of the bispectrum can fit the BAO feature in the measured bispectrum monopole of galaxies. Previous works showed that perturbative models of galaxy bispectrum start failing at the wavenumbers of k ~ 0.1 Mpc/h. We show that when the BAO feature in the bispectrum is separated it can be successfully modeled up to much higher wavenumbers. We validate our modeling on GLAM simulations that were run with and without the BAO feature in the initial conditions. We also quantify the amount of systematic error due to BAO template being offset from the true cosmology. We find that the systematic errors do not exceed 0.3 per cent for reasonable deviations from the true cosmology.
|
astro-ph
|
2,312.04091 |
In 2023, Kuznetsov and Speranski introduced infinitary action logic with multiplexing $!^m\nabla \mathrm{ACT}_\omega$ and proved that the derivability problem for it lies between the $\omega$ and $\omega^\omega$ levels of the hyperarithmetical hierarchy. We prove that this problem is $\Delta^0_{\omega^\omega}$-complete under Turing reductions. Namely, we show that it is recursively isomorphic to the satisfaction predicate for computable infinitary formulas of rank less than $\omega^\omega$ in the language of arithmetic. As a consequence we prove that the closure ordinal for $!^m\nabla \mathrm{ACT}_\omega$ equals $\omega^\omega$. We also prove that the fragment of $!^m\nabla \mathrm{ACT}_\omega$ where Kleene star is not allowed to be in the scope of the subexponential is $\Delta^0_{\omega^\omega}$-complete. Finally, we present a family of logics, which are fragments of $!^m\nabla \mathrm{ACT}_\omega$, such that the complexity of the $k$-th logic lies between $\Delta^0_{\omega^k}$ and $\Delta^0_{\omega^{k+1}}$.
|
cs
|
2,312.03012 |
Networks in machine learning offer examples of complex high-dimensional dynamical systems reminiscent of biological systems. Here, we study the learning dynamics of Generalized Hopfield networks, which permit a visualization of internal memories. These networks have been shown to proceed through a 'feature-to-prototype' transition, as the strength of network nonlinearity is increased, wherein the learned, or terminal, states of internal memories transition from mixed to pure states. Focusing on the prototype learning dynamics of the internal memories we observe a strong resemblance to the canalized, or low-dimensional, dynamics of cells as they differentiate within a Waddingtonian landscape. Dynamically, we demonstrate that learning in a Generalized Hopfield Network proceeds through sequential 'splits' in memory space. Furthermore, order of splitting is interpretable and reproducible. The dynamics between the splits are canalized in the Waddington sense -- robust to variations in detailed aspects of the system. In attempting to make the analogy a rigorous equivalence, we study smaller subsystems that exhibit similar properties to the full system. We combine analytical calculations with numerical simulations to study the dynamical emergence of the feature-to-prototype transition, and the behaviour of splits in the landscape, saddles points, visited during learning. We exhibit regimes where saddles appear and disappear through saddle-node bifurcations, qualitatively changing the distribution of learned memories as the strength of the nonlinearity is varied -- allowing us to systematically investigate the mechanisms that underlie the emergence of Waddingtonian dynamics. Memories can thus differentiate in a predictive and controlled way, revealing new bridges between experimental biology, dynamical systems theory, and machine learning.
|
q-bio
|
2,312.03706 |
Sarcasm, as defined by Merriam-Webster, is the use of words by someone who means the opposite of what he is trying to say. In the field of sentimental analysis of Natural Language Processing, the ability to correctly identify sarcasm is necessary for understanding people's true opinions. Because the use of sarcasm is often context-based, previous research has used language representation models, such as Support Vector Machine (SVM) and Long Short-Term Memory (LSTM), to identify sarcasm with contextual-based information. Recent innovations in NLP have provided more possibilities for detecting sarcasm. In BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, Jacob Devlin et al. (2018) introduced a new language representation model and demonstrated higher precision in interpreting contextualized language. As proposed by Hazarika et al. (2018), CASCADE is a context-driven model that produces good results for detecting sarcasm. This study analyzes a Reddit corpus using these two state-of-the-art models and evaluates their performance against baseline models to find the ideal approach to sarcasm detection.
|
cs
|
2,312.02428 |
Image Retrieval aims to retrieve corresponding images based on a given query. In application scenarios, users intend to express their retrieval intent through various query styles. However, current retrieval tasks predominantly focus on text-query retrieval exploration, leading to limited retrieval query options and potential ambiguity or bias in user intention. In this paper, we propose the Style-Diversified Query-Based Image Retrieval task, which enables retrieval based on various query styles. To facilitate the novel setting, we propose the first Diverse-Style Retrieval dataset, encompassing diverse query styles including text, sketch, low-resolution, and art. We also propose a light-weighted style-diversified retrieval framework. For various query style inputs, we apply the Gram Matrix to extract the query's textural features and cluster them into a style space with style-specific bases. Then we employ the style-init prompt tuning module to enable the visual encoder to comprehend the texture and style information of the query. Experiments demonstrate that our model, employing the style-init prompt tuning strategy, outperforms existing retrieval models on the style-diversified retrieval task. Moreover, style-diversified queries~(sketch+text, art+text, etc) can be simultaneously retrieved in our model. The auxiliary information from other queries enhances the retrieval performance within the respective query.
|
cs
|
2,312.05267 |
In this paper, we show that a possible relationship between the Hubble-Lemaître constant and the universe holographic complexity can be established in the context of a new proposal for the emergence of spacetime, according to which spacetime must emerge from quantum information encoded in quantum correlations without correlate. Such a bridge between the Hubble-Lemaître constant and the universe holographic complexity can shed some light on the issue of the Hubble tension.
|
quant-ph
|
2,312.05442 |
Hyper-luminous infrared galaxies (HyLIRGs) are the most extreme star-forming systems observed in the early Universe, and their properties still elude comprehensive understanding. We have undertaken a large XMM-Newton observing program to probe the total accreting black hole population in three HyLIRGs at z = 2.12, 3.25, and 3.55, gravitationally lensed by foreground galaxies. Selected from the Planck All-Sky Survey to Analyze Gravitationally-lensed Extreme Starbursts (PASSAGES), these HyLIRGs have apparent infrared luminosities > E14 Lsun. Our observations revealed X-ray emission in each of them. PJ1336+49 appears to be dominated by high-mass X-ray binaries (HMXBs). Remarkably, the luminosity of this non-AGN X-ray emission exceeds by a factor of about three the value obtained by calibration with local galaxies with much lower star formation rates. This enhanced X-ray emission most likely highlights the efficacy of dynamical HMXB production within compact clusters, which is an important mode of star formation in HyLIRGs. The remaining two (PJ0116-24 and PJ1053+60) morphologically and spectrally exhibit a compact X-ray component in addition to the extended non-AGN X-ray emission, indicating the presence of Active Galactic Nuclei (AGNs). The AGN appears to be centrally located in the reconstructed source plane images of PJ0116-24, which manifests its star-forming activity predominantly within an extended galactic disk. In contrast, the AGN in the field of PJ1053+60 is projected 60 kpc away from the extreme star-forming galaxy and could be ejected from it. These results underline the synergistic potential of deep X-ray observations with strong lensing for the study of high-energy astrophysical phenomena in HyLIRGs.
|
astro-ph
|
2,312.03283 |
We compute the cohomologies of two strand braid varieties using the two-form present in cluster structures. We confirm these results with proof using Alexander and Poincaré duality. Further, we consider products of braid varieties and their interactions with the cohomologies.
|
math
|
2,312.03043 |
Recent advances in synthetic imaging open up opportunities for obtaining additional data in the field of surgical imaging. This data can provide reliable supplements supporting surgical applications and decision-making through computer vision. Particularly the field of image-guided surgery, such as laparoscopic and robotic-assisted surgery, benefits strongly from synthetic image datasets and virtual surgical training methods. Our study presents an intuitive approach for generating synthetic laparoscopic images from short text prompts using diffusion-based generative models. We demonstrate the usage of state-of-the-art text-to-image architectures in the context of laparoscopic imaging with regard to the surgical removal of the gallbladder as an example. Results on fidelity and diversity demonstrate that diffusion-based models can acquire knowledge about the style and semantics in the field of image-guided surgery. A validation study with a human assessment survey underlines the realistic nature of our synthetic data, as medical personnel detects actual images in a pool with generated images causing a false-positive rate of 66%. In addition, the investigation of a state-of-the-art machine learning model to recognize surgical actions indicates enhanced results when trained with additional generated images of up to 5.20%. Overall, the achieved image quality contributes to the usage of computer-generated images in surgical applications and enhances its path to maturity.
|
cs
|
2,312.02325 |
We present a new crater chronology for Jupiter's Trojan asteroids. This tool can be used to interpret the collisional history of the bodies observed by NASA's Lucy mission. The Lucy mission will visit a total of six Trojan asteroids: Eurybates, Polymele, Orus, Leucus, and the near equal mass binary Patroclus-Menoetius. In addition, Eurybates and Polymele each have a small satellite. Here we present a prediction of Trojan cratering based on current models of how the Solar System and the objects themselves evolved. We give particular emphasis to the time lapsed since their implantation into stable regions near Jupiter's Lagrangian L4 and L4 points. We find that cratering on Trojans is generally dominated by mutual collisions, with the exception of a short period of time (~10 Myr) after implantation, in which cometary impacts may have been significant. For adopted crater scaling laws, we find that the overall spatial density of craters on Trojans is significantly lower than that of Main Belt asteroids on surfaces with similar formation ages. We also discuss specific predictions for similar-sized Eurybates and Orus, and the binary system Patroclus-Menoetius.
|
astro-ph
|
2,312.02937 |
Perception, Planning, and Control form the essential components of autonomy in advanced air mobility. This work advances the holistic integration of these components to enhance the performance and robustness of the complete cyber-physical system. We adapt Perception Simplex, a system for verifiable collision avoidance amidst obstacle detection faults, to the vertical landing maneuver for autonomous air mobility vehicles. We improve upon this system by replacing static assumptions of control capabilities with dynamic confirmation, i.e., real-time confirmation of control limitations of the system, ensuring reliable fulfillment of safety maneuvers and overrides, without dependence on overly pessimistic assumptions. Parameters defining control system capabilities and limitations, e.g., maximum deceleration, are continuously tracked within the system and used to make safety-critical decisions. We apply these techniques to propose a verifiable collision avoidance solution for autonomous aerial mobility vehicles operating in cluttered and potentially unsafe environments.
|
eess
|
2,312.05556 |
Flexoelectricity shows promising applications for self-powered devices with its increased power density. This paper presents a second-order computational homogenization strategy for flexoelectric composite. The macro-micro scale transition, Hill-Mandel energy condition, periodic boundary conditions, and macroscopic constitutive tangents for the two-scale electromechanical coupling are investigated and considered in the homogenization formulation. The macrostructure and microstructure are discretized using $C^1$ triangular finite elements. The second-order multiscale solution scheme is implemented using ABAQUS with user subroutines. Finally, we present numerical examples including parametric analysis of a square plate with holes and the design of piezoelectric materials made of non-piezoelectric materials to demonstrate the numerical implementation and the size-dependent effects of flexoelectricity.
|
cs
|
2,312.06049 |
Global feature based Pedestrian Attribute Recognition (PAR) models are often poorly localized when using Grad-CAM for attribute response analysis, which has a significant impact on the interpretability, generalizability and performance. Previous researches have attempted to improve generalization and interpretation through meticulous model design, yet they often have neglected or underutilized effective prior information crucial for PAR. To this end, a novel Scale and Spatial Priors Guided Network (SSPNet) is proposed for PAR, which is mainly composed of the Adaptive Feature Scale Selection (AFSS) and Prior Location Extraction (PLE) modules. The AFSS module learns to provide reasonable scale prior information for different attribute groups, allowing the model to focus on different levels of feature maps with varying semantic granularity. The PLE module reveals potential attribute spatial prior information, which avoids unnecessary attention on irrelevant areas and lowers the risk of model over-fitting. More specifically, the scale prior in AFSS is adaptively learned from different layers of feature pyramid with maximum accuracy, while the spatial priors in PLE can be revealed from part feature with different granularity (such as image blocks, human pose keypoint and sparse sampling points). Besides, a novel IoU based attribute localization metric is proposed for Weakly-supervised Pedestrian Attribute Localization (WPAL) based on the improved Grad-CAM for attribute response mask. The experimental results on the intra-dataset and cross-dataset evaluations demonstrate the effectiveness of our proposed method in terms of mean accuracy (mA). Furthermore, it also achieves superior performance on the PCS dataset for attribute localization in terms of IoU. Code will be released at .
|
cs
|
2,312.0291 |
We identify rare and visually distinctive galaxy populations by searching for structure within the learned representations of pretrained models. We show that these representations arrange galaxies by appearance in patterns beyond those needed to predict the pretraining labels. We design a clustering approach to isolate specific local patterns, revealing groups of galaxies with rare and scientifically-interesting morphologies.
|
astro-ph
|
2,312.04545 |
The redshift evolution of bars is an important signpost of the dynamic maturity of disk galaxies. To characterize the intrinsic evolution safe from band-shifting effects, it is necessary to gauge how bar properties vary locally as a function of wavelength. We investigate bar properties in 16 nearby galaxies from the Spitzer Survey of Stellar Structure in Galaxies (S4G) at ultraviolet, optical and mid-infrared wavebands. Based on the ellipticity and position angle profiles from fitting elliptical isophotes to the two-dimensional light distribution, we find that both bar length and ellipticity - the latter often used as a proxy for bar strength - increase at bluer wavebands. Bars are 9% longer in the B-band than at 3.6 um. Their ellipticity increases typically by 8% in the B-band, with a significant fraction (>40%) displaying an increase up to 35%. We attribute the increase in bar length to the presence of star forming knots at the end of bars: these regions are brighter in bluer bands, stretching the bar signature further out. The increase in bar ellipticity could be driven by the apparent bulge size: the bulge is less prominent at bluer bands, allowing for thinner ellipses within the bar region. Alternatively, it could be due to younger stellar populations associated to the bar. The resulting effect is that bars appear longer and thinner at bluer wavebands. This indicates that band-shifting effects are significant and need to be corrected for high-redshift studies to reliably gauge any intrinsic evolution of the bar properties with redshift.
|
astro-ph
|
2,312.05176 |
We consider a missing data problem in the context of automatic segmentation methods for Magnetic Resonance Imaging (MRI) brain scans. Usually, automated MRI scan segmentation is based on multiple scans (e.g., T1-weighted, T2-weighted, T1CE, FLAIR). However, quite often a scan is blurry, missing or otherwise unusable. We investigate the question whether a missing scan can be synthesized. We exemplify that this is in principle possible by synthesizing a T2-weighted scan from a given T1-weighted scan. Our first aim is to compute a picture that resembles the missing scan closely, measured by average mean squared error (MSE). We develop/use several methods for this, including a random baseline approach, a clustering-based method and pixel-to-pixel translation method by (Pix2Pix) which is based on conditional GANs. The lowest MSE is achieved by our clustering-based method. Our second aim is to compare the methods with respect to the affect that using the synthesized scan has on the segmentation process. For this, we use a DeepMedic model trained with the four input scan modalities named above. We replace the T2-weighted scan by the synthesized picture and evaluate the segmentations with respect to the tumor identification, using Dice scores as numerical evaluation. The evaluation shows that the segmentation works well with synthesized scans (in particular, with Pix2Pix methods) in many cases.
|
eess
|
2,312.0434 |
In this work, we introduce and construct specific $q$-polynomials that are desired from the well-established families of $q$-orthogonal polynomials, namely little $q$-Jacobi polynomials and $q$-Laguerre polynomials, respectively. We examine these newly constructed $q$-polynomials and observe that they possess integral representations of little $q$-Jacobi polynomials and $q$-Laguerre polynomials. These polynomials solve a third-order $q$-difference equation and display an unconventional four-term recurrence relation. This unique recurrence relation makes us categorize them as $q$-Sobolev-type orthogonal polynomials. This motivation leads to defining the general Sobolev-type orthogonality for $q$-polynomials. Special cases of these polynomials are also explored and discussed. Furthermore, we delve into the behavior of these $q$-orthogonal polynomials of Sobolev type as the parameters approach $1$. We also examine their zeros and interlacing properties.
|
math
|
2,312.05882 |
The purpose of this paper is to analyze the distribution distance between random vectors derived from the magnitude of the analytic wavelet transform of the squared envelopes of Gaussian processes and their large-scale limits. When the Hurst index of the underlying Gaussian process falls below 1/2, the large-scale limit takes the form of a combination of the second and fourth Wiener chaos. In this case, we employ the non-Stein approach to determine the convergence rate in the Kolmogorov metric. In scenarios where the Hurst index of the underlying Gaussian process exceeds 1/2, the large-scale limit is a correlated chi-distributed random process with two degrees of freedom. The convergence rate in the Wasserstein metric is derived through the application of Malliavin calculus and multidimensional Stein's method. Although the large-scale limits are the same for Hurst indices within the interval (1/2,1), notable differences in the upper bounds of convergence rates are observed among Hurst indices falling within the subintervals (1/2,3/4) and (3/4,1).
|
math
|
2,312.02367 |
Our goals fundamentally shape how we experience the world. For example, when we are hungry, we tend to view objects in our environment according to whether or not they are edible (or tasty). Alternatively, when we are cold, we may view the very same objects according to their ability to produce heat. Computational theories of learning in cognitive systems, such as reinforcement learning, use the notion of "state-representation" to describe how agents decide which features of their environment are behaviorally-relevant and which can be ignored. However, these approaches typically assume "ground-truth" state representations that are known by the agent, and reward functions that need to be learned. Here we suggest an alternative approach in which state-representations are not assumed veridical, or even pre-defined, but rather emerge from the agent's goals through interaction with its environment. We illustrate this novel perspective by inferring the goals driving rat behavior in an odor-guided choice task and discuss its implications for developing, from first principles, an information-theoretic account of goal-directed state representation learning and behavior.
|
math
|
2,312.04429 |
Text-to-image generation using diffusion models has seen explosive popularity owing to their ability in producing high quality images adhering to text prompts. However, production-grade diffusion model serving is a resource intensive task that not only require high-end GPUs which are expensive but also incurs considerable latency. In this paper, we introduce a technique called approximate-caching that can reduce such iterative denoising steps for an image generation based on a prompt by reusing intermediate noise states created during a prior image generation for similar prompts. Based on this idea, we present an end to end text-to-image system, Nirvana, that uses the approximate-caching with a novel cache management-policy Least Computationally Beneficial and Frequently Used (LCBFU) to provide % GPU compute savings, 19.8% end-to-end latency reduction and 19% dollar savings, on average, on two real production workloads. We further present an extensive characterization of real production text-to-image prompts from the perspective of caching, popularity and reuse of intermediate states in a large production environment.
|
cs
|
2,312.03335 |
Detecting non-termination is crucial for ensuring program correctness and security, such as preventing denial-of-service attacks. While termination analysis has been studied for many years, existing methods have limited scalability and are only effective on small programs. To address this issue, we propose a practical termination checking technique, called EndWatch, for detecting non-termination caused by infinite loops through testing. Specifically, we introduce two methods to generate non-termination oracles based on checking state revisits, i.e., if the program returns to a previously visited state at the same program location, it does not terminate. The non-termination oracles can be incorporated into testing tools (e.g., AFL used in this paper) to detect non-termination in large programs. For linear loops, we perform symbolic execution on individual loops to infer State Revisit Conditions (SRCs) and instrument SRCs into target loops. For non-linear loops, we instrument target loops for checking concrete state revisits during execution. We evaluated EndWatch on standard benchmarks with small-sized programs and real-world projects with large-sized programs. The evaluation results show that EndWatch is more effective than the state-of-the-art tools on standard benchmarks (detecting 87% of non-terminating programs while the best baseline detects only 67%), and useful in detecting non-termination in real-world projects (detecting 90% of known non-termination CVEs and 4 unknown bugs).
|
cs
|
2,312.03867 |
Machine learning (ML) models used in prediction and classification tasks may display performance disparities across population groups determined by sensitive attributes (e.g., race, sex, age). We consider the problem of evaluating the performance of a fixed ML model across population groups defined by multiple sensitive attributes (e.g., race and sex and age). Here, the sample complexity for estimating the worst-case performance gap across groups (e.g., the largest difference in error rates) increases exponentially with the number of group-denoting sensitive attributes. To address this issue, we propose an approach to test for performance disparities based on Conditional Value-at-Risk (CVaR). By allowing a small probabilistic slack on the groups over which a model has approximately equal performance, we show that the sample complexity required for discovering performance violations is reduced exponentially to be at most upper bounded by the square root of the number of groups. As a byproduct of our analysis, when the groups are weighted by a specific prior distribution, we show that Rényi entropy of order $2/3$ of the prior distribution captures the sample complexity of the proposed CVaR test algorithm. Finally, we also show that there exists a non-i.i.d. data collection strategy that results in a sample complexity independent of the number of groups.
|
math
|
2,312.05982 |
The paper presents an approach for overcoming modeling problems of typical life science applications with partly unknown mechanisms and lacking quantitative data: A model family of reaction diffusion equations is built up on a mesoscopic scale and uses classes of feasible functions for reaction and taxis terms. The classes are found by translating biological knowledge into mathematical conditions and the analysis of the models further constrains the classes. Numerical simulations allow comparing single models out of the model family with available qualitative information on the solutions from observations. The method provides insight into a hierarchical order of the mechanisms. The method is applied to the clinics for liver inflammation such as metabolic dysfunction-associated steatohepatitis (MASH) or viral hepatitis where reasons for the chronification of disease are still unclear and time- and space-dependent data is unavailable.
|
q-bio
|
2,312.04043 |
In this paper, we democratise 3D content creation, enabling precise generation of 3D shapes from abstract sketches while overcoming limitations tied to drawing skills. We introduce a novel part-level modelling and alignment framework that facilitates abstraction modelling and cross-modal correspondence. Leveraging the same part-level decoder, our approach seamlessly extends to sketch modelling by establishing correspondence between CLIPasso edgemaps and projected 3D part regions, eliminating the need for a dataset pairing human sketches and 3D shapes. Additionally, our method introduces a seamless in-position editing process as a byproduct of cross-modal part-aligned modelling. Operating in a low-dimensional implicit space, our approach significantly reduces computational demands and processing time.
|
cs
|
2,312.02747 |
In this article, we study the dynamic variant and the equilibrium variant of the model of Rydberg atoms on a square ladder. In the dynamic case, we obtain the jamming limit for all values of $b \ge 1$, where $b$ represents the so-called blockade range of a Rydberg atom. In the equilibrium case, we derive the complexity function for all values of $b \ge 1$. By comparing these results, we highlight significant differences in the behavior of the two models as $b$ approaches infinity.
|
math
|
2,312.03152 |
This is a model-independent analysis that investigates the statistical isotropy in the Local Universe using the ALFALFA survey data ($0 < z < 0.06$). We investigate the angular distribution of HI extra-galactic sources from the ALFALFA catalogue and study whether they are compatible with the statistical isotropy hypothesis using the two-point angular correlation function (2PACF). Aware that the Local Universe is plenty of clustered structures and large voids, we compute the 2PACF with the Landy-Szalay estimator performing directional analyses to inspect 10 sky regions. We investigate these 2PACF using power-law best-fit analyses, and determine the statistical significance of the best-fit parameters for the 10 ALFALFA regions by comparison with the ones obtained through the same procedure applied to a set of mock catalogues produced under the homogeneity and isotropy hypotheses. Our conclusion is that the Local Universe, as mapped by the HI sources of the ALFALFA survey, is in agreement with the hypothesis of statistical isotropy within $2\,\sigma$ confidence level, for small and large angle analyses, with the only exception of one region -- located near the Dipole Repeller -- which appears slightly outlier ($2.4\,\sigma$). Interestingly, regarding the large angular distribution of the HI sources, we found 3 regions where the presence of cosmic voids reported in the literature left their signature in our 2PACF, suggesting projected large underdensities there, with number-density contrast $\delta \simeq -0.7$. According to the current literature these regions correspond, partially, to the sky position of the void structures known as Local Cosmic Void and Dipole Repeller.
|
astro-ph
|
2,312.04659 |
Dimensions of level sets of generic continuous functions and generic Hölder functions defined on a fractal $F$ encode information about the geometry, ``the thickness" of $F$. While in the continuous case this quantity is related to a reasonably tame dimension notion which is called the topological Hausdorff dimension of $F$, the Hölder case seems to be highly nontrivial. A number of earlier papers attempted to deal with this problem, carrying out investigation in the case of Hausdorff dimension and box dimension. In this paper we continue our study of the Hausdorff dimension of almost every level set of generic $1$-Hölder-$ {\alpha}$ functions, denoted by $D_{*}( {\alpha}, F)$. We substantially improve previous lower and upper bounds on $D_{*}( {\alpha}, \Delta)$, where $\Delta$ is the Sierpiński triangle, achieving asymptotically equal bounds as $\alpha\to 0+$. Using a similar argument, we also give an even stronger lower bound on the generic lower box dimension of level sets. Finally, we construct a connected fractal $F$ on which there is a phase transition of $D_{*}( {\alpha}, F)$, thus providing the first example exhibiting this behavior.
|
math
|
2,312.02454 |
Two near-threshold peaking structures with spin-parities $J^{PC}=0^{++}$ were recently discovered by the LHCb Collaboration in the $D_s^+D_s^-$ invariant mass distribution of the decay process $B^+\to D_s^+D_s^-K^+$. In our study, we employed a coupled-channel model to fit the experimental results published by the LHCb collaboration, simultaneously fitting the model to the invariant mass distributions of $M_{D_s^+D_s^-}$, $M_{D_s^+K^+}$, and $M_{D_s^-K^+}$. We utilized a coupled-channel model to search for the poles of $X(3960)$ and $X_0(4140)$. The determination of the poles is meaningful in itself, and it also lays an foundation for the future research on $X(3960)$ and $X_0(4140)$. Upon turning off the coupled-channel and performing another fit, we observed a change in the fitting quality, the effect was almost entirely due to the peak of $X(3960)$, so we suggest that $X(3960)$ may not be a kinematic effect.
|
hep-ex
|
2,312.04022 |
Reshaping, a point operation that alters the characteristics of signals, has been shown capable of improving the compression ratio in video coding practices. Out-of-loop reshaping that directly modifies the input video signal was first adopted as the supplemental enhancement information~(SEI) for the HEVC/H.265 without the need of altering the core design of the video codec. VVC/H.266 further improves the coding efficiency by adopting in-loop reshaping that modifies the residual signal being processed in the hybrid coding loop. In this paper, we theoretically analyze the rate-distortion performance of the in-loop reshaping and use experiments to verify the theoretical result. We prove that the in-loop reshaping can improve coding efficiency when the entropy coder adopted in the coding pipeline is suboptimal, which is in line with the practical scenarios that video codecs operate in. We derive the PSNR gain in a closed form and show that the theoretically predicted gain is consistent with that measured from experiments using standard testing video sequences.
|
eess
|
2,312.02305 |
Virtual and mixed reality (VR, MR) technologies offer a powerful solution for on-the-ground flight training curricula. While these technologies offer safer and cheaper instructional programs, it is still unclear how they impact neuronal brain dynamics. Indeed, MR simulations engage students in a strange mix of incongruous visual, somatosensory and vestibular sensory input. Characterizing brain dynamics during MR simulation is important for understanding cognitive processes during virtual flight training. To this end, we studies the delays introduced in the neuronal stream from the retina to the visual cortex when presented with visual stimuli using a Varjo-XR3 headset. We recorded cortical visual evoked potentials (VEPs) from 6 subjects under two conditions. First, we recorded normal VEPs triggered by short flashes. Second, we recorded VEPs triggered by an internal image of the flashes produced by the Varjo-XR3 headset. All subjects had used the headset before and were familiar with immersive experiences. Our results show mixed-reality stimulation imposes a small, but consistent, 4 [ms] processing delay in the N2-VEP component during MR stimulation as compared to direct stimulation. Also we found that VEP amplitudes during MR stimulation were also decreased. These results suggest that visual cognition during mixed-reality training is delayed, not only by the unavoidalbe hardware/software processing delays of the headset and the attached computer, but also by an extra biological delay induced by the headset's limited visual display in terms of the image intensity and contrast. As flight training is a demanding task, this study measures visual signal latency to better understand how MR affects the sensation of immersion.
|
q-bio
|
2,312.02914 |
In this work, we tackle the problem of unsupervised domain adaptation (UDA) for video action recognition. Our approach, which we call UNITE, uses an image teacher model to adapt a video student model to the target domain. UNITE first employs self-supervised pre-training to promote discriminative feature learning on target domain videos using a teacher-guided masked distillation objective. We then perform self-training on masked target data, using the video student model and image teacher model together to generate improved pseudolabels for unlabeled target videos. Our self-training process successfully leverages the strengths of both models to achieve strong transfer performance across domains. We evaluate our approach on multiple video domain adaptation benchmarks and observe significant improvements upon previously reported results.
|
cs
|
2,312.04561 |
We offer a new perspective on approaching the task of video generation. Instead of directly synthesizing a sequence of frames, we propose to render a video by warping one static image with a generative deformation field (GenDeF). Such a pipeline enjoys three appealing advantages. First, we can sufficiently reuse a well-trained image generator to synthesize the static image (also called canonical image), alleviating the difficulty in producing a video and thereby resulting in better visual quality. Second, we can easily convert a deformation field to optical flows, making it possible to apply explicit structural regularizations for motion modeling, leading to temporally consistent results. Third, the disentanglement between content and motion allows users to process a synthesized video through processing its corresponding static image without any tuning, facilitating many applications like video editing, keypoint tracking, and video segmentation. Both qualitative and quantitative results on three common video generation benchmarks demonstrate the superiority of our GenDeF method.
|
cs
|
2,312.05931 |
We construct a form of the $D_4^+$-singularity of fronts in $R^3$ which uses coordinate transformation on the source and isometry on the target. As an application, we calculate differential geometric invariants near the $D_4^+$-singularity, and give a Gauss-Bonnet type theorem for fronts allowing to have this singularity.
|
math
|
2,312.02345 |
With the goal of understanding the visual concepts that CLIP associates with text prompts, we show that the latent space of CLIP can be visualized solely in terms of linear transformations on simple geometric primitives like circles and straight lines. Although existing approaches achieve this by sketch-synthesis-through-optimization, they do so on the space of Bézier curves, which exhibit a wastefully large set of structures that they can evolve into, as most of them are non-essential for generating meaningful sketches. We present CLIPDrawX, an algorithm that provides significantly better visualizations for CLIP text embeddings, using only simple primitive shapes like straight lines and circles. This constrains the set of possible outputs to linear transformations on these primitives, thereby exhibiting an inherently simpler mathematical form. The synthesis process of CLIPDrawX can be tracked end-to-end, with each visual concept being explained exclusively in terms of primitives. Implementation will be released upon acceptance. Project Page: $\href{}{\text{}}$.
|
cs
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.