Search

Breast cancer is a disease in which cells in the breast grow out of control. After surgery, chemotherapy, and other invasive treatments, hyperthermia is a suitable choice with the minimal side effects. In this paper, the treatment of breast cancer using a combination of laser and ultrasound irradiation, in the presence of gold nanoparticles, is investigated. In the simulations, the breast tissue is represented as a multilayer structure and the tumor is supposed to consist two parts: a superficial section and a deeper region. In the initial stage, the superficial parts of the tumor, which also contain gold nanoparticles, are exposed to a continuous laser for 50 seconds followed by a cooling period of 20 seconds. Then, for deeper sections, ultrasound irradiation is utilized. The results indicated that the tissue necrosis volume was enhanced by the application of nanoparticles. Furthermore, it was demonstrated that the combinational application of both of laser and ultrasound irradiation could eradicate both the superficial parts of the tumor and the deep parts.

Generalized supergravity equations (GSEs) were originally proposed as a modification of the standard IIB supergravity equations to satisfy the background of $\eta$-deformed $AdS_5 \times S^5$ introduced by Arutyunov {\it et al}. In this study, we proceed to write down the GSEs for the Wess-Zumino-Witten (WZW) models based on Lie groups. First, we simplify the GSEs for the WZW model by imposing the conditions for vanishing of the one-loop beta function equations. Then, by introducing an {\it Ansatz} for the Killing vector field $I$, it is shown that the Killing equation ${\cal L}_{_{I}} G_{_{\mu \nu}}=0$ is held. In this way, we solve the simplified GSEs for the WZW models constructed on Lie groups up to dimension five., Comment: 14 pages, one table

Conditional Flow Matching (CFM) models can generate high-quality samples from a non-informative prior, but they can be slow, often needing hundreds of network evaluations (NFE). To address this, we propose Implicit Dynamical Flow Fusion (IDFF); IDFF learns a new vector field with an additional momentum term that enables taking longer steps during sample generation while maintaining the fidelity of the generated distribution. Consequently, IDFFs reduce the NFEs by a factor of ten (relative to CFMs) without sacrificing sample quality, enabling rapid sampling and efficient handling of image and time-series data generation tasks. We evaluate IDFF on standard benchmarks such as CIFAR-10 and CelebA for image generation. We achieved likelihood and quality performance comparable to CFMs and diffusion-based models with fewer NFEs. IDFF also shows superior performance on time-series datasets modeling, including molecular simulation and sea surface temperature (SST) datasets, highlighting its versatility and effectiveness across different domains.

Recent advancements in predicting pedestrian crossing intentions for Autonomous Vehicles using Computer Vision and Deep Neural Networks are promising. However, the black-box nature of DNNs poses challenges in understanding how the model works and how input features contribute to final predictions. This lack of interpretability delimits the trust in model performance and hinders informed decisions on feature selection, representation, and model optimisation; thereby affecting the efficacy of future research in the field. To address this, we introduce Context-aware Permutation Feature Importance (CAPFI), a novel approach tailored for pedestrian intention prediction. CAPFI enables more interpretability and reliable assessments of feature importance by leveraging subdivided scenario contexts, mitigating the randomness of feature values through targeted shuffling. This aims to reduce variance and prevent biased estimations in importance scores during permutations. We divide the Pedestrian Intention Estimation (PIE) dataset into 16 comparable context sets, measure the baseline performance of five distinct neural network architectures for intention prediction in each context, and assess input feature importance using CAPFI. We observed nuanced differences among models across various contextual characteristics. The research reveals the critical role of pedestrian bounding boxes and ego-vehicle speed in predicting pedestrian intentions, and potential prediction biases due to the speed feature through cross-context permutation evaluation. We propose an alternative feature representation by considering proximity change rate for rendering dynamic pedestrian-vehicle locomotion, thereby enhancing the contributions of input features to intention prediction. These findings underscore the importance of contextual features and their diversity to develop accurate and robust intent-predictive models.

Dark matter, an important portion of compact objects, can influence different phenomena in neutron stars. The spontaneous scalarization in the scalar-tensor gravity has been proposed for neutron stars. Here, we investigate the spontaneous scalarization in dark matter admixed neutron stars. Applying the dark matter equations of state, we calculate the structure of scalarized neutron stars containing dark matter. The dark matter equations of state are based on observational data from the rotational curves of galaxies and the fermionic self-interacting dark matter. Our results verify that the spontaneous scalarization is affected by the dark matter pressure in neutron stars. Depending on the central density of scalarized dark matter admixed neutron stars, the dark matter pressure alters the central scalar field. The increase of dark matter pressure in low-density scalarized stars amplifies the central scalar field. However, the pressure of dark matter in high-density scalarized stars suppresses the central scalar field. Our calculations confirm that the stars in the merger event GW170817 and in the low-mass X-ray binary 4U 1820-30 can be scalarized dark matter admixed neutron stars., Comment: 23 pages, 11 figures. Accepted for publication in European Physical Journal C

High-resolution solar observations have revealed the existence of small-scale vortices, as seen in chromospheric intensity maps and velocity diagnostics. Frequently, these vortices have been observed near magnetic flux concentrations, indicating a link between swirls and the evolution of the small-scale magnetic fields. Vortices have also been studied with magneto-hydrodynamic (MHD) numerical simulations of the solar atmosphere, revealing their complexity, dynamics, and magnetic nature. In particular, it has been proposed that a rotating magnetic field structure driven by a photospheric vortex flow at its footprint produces the chromospheric swirling plasma motion. We present a complete and comprehensive description of the time evolution of a small-scale magnetic flux concentration interacting with the intergranular vortex flow and affected by processes of intensification and weakening of its magnetic field. In addition, we study the chromospheric dynamics associated with the interaction, including the analysis of a chromospheric swirl and an impulsive chromospheric jet., Comment: Accepted for publication A&A

Adverse conditions like snow, rain, nighttime, and fog, pose challenges for autonomous driving perception systems. Existing methods have limited effectiveness in improving essential computer vision tasks, such as semantic segmentation, and often focus on only one specific condition, such as removing rain or translating nighttime images into daytime ones. To address these limitations, we propose a method to improve the visual quality and clarity degraded by such adverse conditions. Our method, AllWeather-Net, utilizes a novel hierarchical architecture to enhance images across all adverse conditions. This architecture incorporates information at three semantic levels: scene, object, and texture, by discriminating patches at each level. Furthermore, we introduce a Scaled Illumination-aware Attention Mechanism (SIAM) that guides the learning towards road elements critical for autonomous driving perception. SIAM exhibits robustness, remaining unaffected by changes in weather conditions or environmental scenes. AllWeather-Net effectively transforms images into normal weather and daytime scenes, demonstrating superior image enhancement results and subsequently enhancing the performance of semantic segmentation, with up to a 5.3% improvement in mIoU in the trained domain. We also show our model's generalization ability by applying it to unseen domains without re-training, achieving up to 3.9% mIoU improvement. Code can be accessed at: https://github.com/Jumponthemoon/AllWeatherNet.

Human understandable explanation of deep learning models is necessary for many critical and sensitive applications. Unlike image or tabular data where the importance of each input feature (for the classifier's decision) can be directly projected into the input, time series distinguishable features (e.g. dominant frequency) are often hard to manifest in time domain for a user to easily understand. Moreover, most explanation methods require a baseline value as an indication of the absence of any feature. However, the notion of lack of feature, which is often defined as black pixels for vision tasks or zero/mean values for tabular data, is not well-defined in time series. Despite the adoption of explainable AI methods (XAI) from tabular and vision domain into time series domain, these differences limit the application of these XAI methods in practice. In this paper, we propose a simple yet effective method that allows a model originally trained on time domain to be interpreted in other explanation spaces using existing methods. We suggest four explanation spaces that each can potentially alleviate these issues in certain types of time series. Our method can be readily adopted in existing platforms without any change to trained models or XAI methods. The code is available at https://github.com/shrezaei/TS-X-spaces.

Explainable Artificial Intelligence (XAI) plays a crucial role in fostering transparency and trust in AI systems, where traditional XAI approaches typically offer one level of abstraction for explanations, often in the form of heatmaps highlighting single or multiple input features. However, we ask whether abstract reasoning or problem-solving strategies of a model may also be relevant, as these align more closely with how humans approach solutions to problems. We propose a framework, called Symbolic XAI, that attributes relevance to symbolic queries expressing logical relationships between input features, thereby capturing the abstract reasoning behind a model's predictions. The methodology is built upon a simple yet general multi-order decomposition of model predictions. This decomposition can be specified using higher-order propagation-based relevance methods, such as GNN-LRP, or perturbation-based explanation methods commonly used in XAI. The effectiveness of our framework is demonstrated in the domains of natural language processing (NLP), vision, and quantum chemistry (QC), where abstract symbolic domain knowledge is abundant and of significant interest to users. The Symbolic XAI framework provides an understanding of the model's decision-making process that is both flexible for customization by the user and human-readable through logical formulas.

We construct a new class of $(n+1)$-dimensional Lifshitz dilaton black brane solutions in the presence of the cubic quasitopological gravity for a flat boundary. The related action supports asymptotically Lifshitz solutions by applying some conditions which are used throughout the paper. We have to add a new boundary term and some new counterterms to the bulk action to have finite solutions. Then we define a finite stress tensor complex by which we can calculate the energy density of the quasitopological Lifshitz dilaton black brane. It is not possible to obtain analytical solutions, and so we use some expantions to probe the functions behaviors near the horizon and at the infinity. Combining the equations, we can attain a total constant along the coordinate $r$. At the horizon, this constant is proportional to the product of the temperature and the entropy and at the infinity, the total constant shows the energ density of the quasitopological Lifshitz dilaton black brane. Therefore, we can reach to a relation between the conserved quantities temperature, entropy and the energy density and get a smarr-type formula. Using the first law of thermodynamics, we can find a relation between the entropy and the temperature and then ontain the heat capacity. Our results show that the quasitopological Lifshitz dilaton black brane solutions are thermally stable for each positive values of the dynamical critiacl exponent, $z$., Comment: 14 pages and 2 figures

The popularity of deep learning methods in the time series domain boosts interest in interpretability studies, including counterfactual (CF) methods. CF methods identify minimal changes in instances to alter the model predictions. Despite extensive research, no existing work benchmarks CF methods in the time series domain. Additionally, the results reported in the literature are inconclusive due to the limited number of datasets and inadequate metrics. In this work, we redesign quantitative metrics to accurately capture desirable characteristics in CFs. We specifically redesign the metrics for sparsity and plausibility and introduce a new metric for consistency. Combined with validity, generation time, and proximity, we form a comprehensive metric set. We systematically benchmark 6 different CF methods on 20 univariate datasets and 10 multivariate datasets with 3 different classifiers. Results indicate that the performance of CF methods varies across metrics and among different models. Finally, we provide case studies and a guideline for practical usage., Comment: 15 pages, 27 figures

With generative models producing high quality images that are indistinguishable from real ones, there is growing concern regarding the malicious usage of AI-generated images. Imperceptible image watermarking is one viable solution towards such concerns. Prior watermarking methods map the image to a latent space for adding the watermark. Moreover, Latent Diffusion Models (LDM) generate the image in the latent space of a pre-trained autoencoder. We argue that this latent space can be used to integrate watermarking into the generation process. To this end, we present LaWa, an in-generation image watermarking method designed for LDMs. By using coarse-to-fine watermark embedding modules, LaWa modifies the latent space of pre-trained autoencoders and achieves high robustness against a wide range of image transformations while preserving perceptual quality of the image. We show that LaWa can also be used as a general image watermarking method. Through extensive experiments, we demonstrate that LaWa outperforms previous works in perceptual quality, robustness against attacks, and computational complexity, while having very low false positive rate. Code is available here., Comment: Accepted to ECCV 2024

Monolayer NiI$_2$ garners large research interest due to its multiferroic behavior stemming from the interplay between its non-collinear magnetic order and the spin-orbit coupling. This prompts an investigation into the stability of the magnetic order in NiI$_2$ and similar materials under external stimuli. In this work, we report the effect of biaxial and uniaxial strain on the magnetic ground state, the critical temperature, and the magnonic properties of the NiX$_2$ (X = I, Br, Cl) monolayers. For all three materials, we reveal intricate strain-dependent phase diagrams, including ferromagnetic, helimagnetic, and skyrmionic phases. Moreover, we discuss the necessity of considering the biquadratic exchange interaction in the latter analysis. We reveal that the biquadratic exchange significantly alters both the magnetic ground state and the critical temperature of the magnetic order, and we demonstrate that its importance becomes even more explicit when monolayer Ni-dihalides are strained. Finally, we calculate the magnonic dispersion for the predicted magnetic states, showing that the skyrmionic phase functions as a magnonic crystal, and demonstrate the presence of strain-tunable soft magnon modes at finite wavevectors in the helimagnetic phase.

A spatiotemporal deep learning framework is proposed that is capable of 2D full-field prediction of fracture in concrete mesostructures. This framework not only predicts fractures but also captures the entire history of the fracture process, from the crack initiation in the interfacial transition zone to the subsequent propagation of the cracks in the mortar matrix. In addition, a convolutional neural network is developed which can predict the averaged stress-strain curve of the mesostructures. The UNet modeling framework, which comprises an encoder-decoder section with skip connections, is used as the deep learning surrogate model. Training and test data are generated from high-fidelity fracture simulations of randomly generated concrete mesostructures. These mesostructures include geometric variabilities such as different aggregate particle geometrical features, spatial distribution, and the total volume fraction of aggregates. The fracture simulations are carried out in Abaqus, utilizing the cohesive phase-field fracture modeling technique as the fracture modeling approach. In this work, to reduce the number of training datasets, the spatial distribution of three sets of material properties for three-phase concrete mesostructures, along with the spatial phase-field damage index, are fed to the UNet to predict the corresponding stress and spatial damage index at the subsequent step. It is shown that after the training process using this methodology, the UNet model is capable of accurately predicting damage on the unseen test dataset by using 470 datasets. Moreover, another novel aspect of this work is the conversion of irregular finite element data into regular grids using a developed pipeline. This approach allows for the implementation of less complex UNet architecture and facilitates the integration of phase-field fracture equations into surrogate models for future developments.

In recent years, there has been increased demand for speech-to-speech translation (S2ST) systems in industry settings. Although successfully commercialized, cloning-based S2ST systems expose their distributors to liabilities when misused by individuals and can infringe on personality rights when exploited by media organizations. This work proposes a regulated S2ST framework called Preset-Voice Matching (PVM). PVM removes cross-lingual voice cloning in S2ST by first matching the input voice to a similar prior consenting speaker voice in the target-language. With this separation, PVM avoids cloning the input speaker, ensuring PVM systems comply with regulations and reduce risk of misuse. Our results demonstrate PVM can significantly improve S2ST system run-time in multi-speaker settings and the naturalness of S2ST synthesized speech. To our knowledge, PVM is the first explicitly regulated S2ST framework leveraging similarly-matched preset-voices for dynamic S2ST tasks., Comment: Accepted to the ACL PrivateNLP 2024 Workshop, 7 pages, 2 figures

We present results from a search for X-ray/gamma-ray counterparts of gravitational-wave (GW) candidates from the third observing run (O3) of the LIGO-Virgo-KAGRA (LVK) network using the Swift Burst Alert Telescope (Swift-BAT). The search includes 636 GW candidates received in low latency, 86 of which have been confirmed by the offline analysis and included in the third cumulative Gravitational-Wave Transient Catalogs (GWTC-3). Targeted searches were carried out on the entire GW sample using the maximum--likelihood NITRATES pipeline on the BAT data made available via the GUANO infrastructure. We do not detect any significant electromagnetic emission that is temporally and spatially coincident with any of the GW candidates. We report flux upper limits in the 15-350 keV band as a function of sky position for all the catalog candidates. For GW candidates where the Swift-BAT false alarm rate is less than 10$^{-3}$ Hz, we compute the GW--BAT joint false alarm rate. Finally, the derived Swift-BAT upper limits are used to infer constraints on the putative electromagnetic emission associated with binary black hole mergers., Comment: 50 pages, 10 figures, 4 tables

Blood biomarkers are an essential tool for healthcare providers to diagnose, monitor, and treat a wide range of medical conditions. Current reference values and recommended ranges often rely on population-level statistics, which may not adequately account for the influence of inter-individual variability driven by factors such as lifestyle and genetics. In this work, we introduce a novel framework for predicting future blood biomarker values and define personalized references through learned representations from lifestyle data (physical activity and sleep) and blood biomarkers. Our proposed method learns a similarity-based embedding space that captures the complex relationship between biomarkers and lifestyle factors. Using the UK Biobank (257K participants), our results show that our deep-learned embeddings outperform traditional and current state-of-the-art representation learning techniques in predicting clinical diagnosis. Using a subset of UK Biobank of 6440 participants who have follow-up visits, we validate that the inclusion of these embeddings and lifestyle factors directly in blood biomarker models improves the prediction of future lab values from a single lab visit. This personalized modeling approach provides a foundation for developing more accurate risk stratification tools and tailoring preventative care strategies. In clinical settings, this translates to the potential for earlier disease detection, more timely interventions, and ultimately, a shift towards personalized healthcare.

We introduce a method that combines neural operators, physics-informed machine learning, and standard numerical methods for solving PDEs. The proposed approach extends each of the aforementioned methods and unifies them within a single framework. We can parametrically solve partial differential equations in a data-free manner and provide accurate sensitivities, meaning the derivatives of the solution space with respect to the design space. These capabilities enable gradient-based optimization without the typical sensitivity analysis costs, unlike adjoint methods that scale directly with the number of response functions. Our Finite Operator Learning (FOL) approach uses an uncomplicated feed-forward neural network model to directly map the discrete design space (i.e. parametric input space) to the discrete solution space (i.e. finite number of sensor points in the arbitrary shape domain) ensuring compliance with physical laws by designing them into loss functions. The discretized governing equations, as well as the design and solution spaces, can be derived from any well-established numerical techniques. In this work, we employ the Finite Element Method (FEM) to approximate fields and their spatial derivatives. Subsequently, we conduct Sobolev training to minimize a multi-objective loss function, which includes the discretized weak form of the energy functional, boundary conditions violations, and the stationarity of the residuals with respect to the design variables. Our study focuses on the steady-state heat equation within heterogeneous materials that exhibits significant phase contrast and possibly temperature-dependent conductivity. The network's tangent matrix is directly used for gradient-based optimization to improve the microstructure's heat transfer characteristics. ..., Comment: arXiv admin note: text overlap with arXiv:2401.02363

The Wide Field Imager (WFI), one of two instruments on ESA's next large X-ray mission Athena, is designed for imaging spectroscopy of X-rays in the range of 0.2 to 15 keV with a large field of view and high count rate capability. The focal plane consists of back-illuminated DEPFET (Depleted p-channel field effect transistor) sensors that have a high radiation tolerance and provide a near Fano-limited energy resolution. To achieve this, a very low noise readout is required, about 3 electrons ENC at beginning of life is foreseen. This makes the device very susceptible to any radiation induced worsening of the readout noise. The main mechanism of degradation will be the increase of dark current due to displacement damage caused primarily by high energy protons. To study the expected performance degradation, a prototype detector module with fully representative pixel layout and fabrication technology was irradiated with 62.4 MeV protons at the accelerator facility MedAustron in Wiener Neustadt. A total dose equivalent to 3.3 $\textstyle{10^{9}}$ 10-MeV protons/$\mathrm{cm^{2}}$ was applied in two steps. During, in-between and after the irradiations the detector remained at the operating temperature of 213 K and was fully biased and operated. Data was recorded to analyze the signal of all incident particles. We report on the increase of dark current after the irradiation and present the current related damage rate at 213 K. The effect of low temperature annealing at 213 K , 236 K, 253 K, 273 K, and 289 K is presented., Comment: to be published in SPIE Proceeding of Astronomical Telescopes + Instrumentation, 2024

In the pursuit of advancing solar energy technologies, this study presents 20 direct and quasi-direct band gap silicon crystalline semiconductors that satisfy the Shockley-Queisser limit, a benchmark for solar cell efficiency. Employing two evolutionary algorithm-based searches, we optimize structures and calculate fitness function using the DFTB method and Gaussian approximation potential. Following the preselection of structures based on energy considerations, we further optimize them using PBEsol DFT. Subsequently, we screen the structures based on their band gap, employing a DFTB method tailored for band gap calculation of silicon crystals. To ensure accurate band gap determination, we employ HSE and GW methods. To validate the structural stability, we employ phonon analysis via linear regression algorithm applied to PBEsol DFT data. Significantly, the structures unveiled in this study are of great importance due to their proven stability from both mechanical and dynamic perspectives. Furthermore, the ductility and low density of certain structures enhance their potential application. We examine the optical properties by studying the imaginary part of the dielectric function by solving the Bethe-Salpeter Equation on top of GW approximation. By calculating the SLME, we achieve an efficiency of 32.7% for Si$_{22}$ at a thickness of 500 nm. Moreover, the study harnesses various machine learning algorithms to develop a predictive model for the band gap energy of these silicon structures. Input data for machine learning models are derived from structural MBTR and SOAP descriptors, as well as DFT outputs. Notably, the results reveal that features extracted from DFT outperform the MBTR and SOAP descriptors., Comment: 14 pages, 8 figures, 5 tables

This paper presents the SLEGO (Software-Lego) system, a collaborative analytics platform that bridges the gap between experienced developers and novice users using a cloud-based platform with modular, reusable microservices. These microservices enable developers to share their analytical tools and workflows, while a simple graphical user interface (GUI) allows novice users to build comprehensive analytics pipelines without programming skills. Supported by a knowledge base and a Large Language Model (LLM) powered recommendation system, SLEGO enhances the selection and integration of microservices, increasing the efficiency of analytics pipeline construction. Case studies in finance and machine learning illustrate how SLEGO promotes the sharing and assembly of modular microservices, significantly improving resource reusability and team collaboration. The results highlight SLEGO's role in democratizing data analytics by integrating modular design, knowledge bases, and recommendation systems, fostering a more inclusive and efficient analytical environment., Comment: 11 pages, 10 figures, 5 tables

In the field of computational advertising, the integration of ads into the outputs of large language models (LLMs) presents an opportunity to support these services without compromising content integrity. This paper introduces novel auction mechanisms for ad allocation and pricing within the textual outputs of LLMs, leveraging retrieval-augmented generation (RAG). We propose a segment auction where an ad is probabilistically retrieved for each discourse segment (paragraph, section, or entire output) according to its bid and relevance, following the RAG framework, and priced according to competing bids. We show that our auction maximizes logarithmic social welfare, a new notion of welfare that balances allocation efficiency and fairness, and we characterize the associated incentive-compatible pricing rule. These results are extended to multi-ad allocation per segment. An empirical evaluation validates the feasibility and effectiveness of our approach over several ad auction scenarios, and exhibits inherent tradeoffs in metrics as we allow the LLM more flexibility to allocate ads.

Recent sequence modeling approaches using Selective State Space Sequence Models, referred to as Mamba models, have seen a surge of interest. These models allow efficient processing of long sequences in linear time and are rapidly being adopted in a wide range of applications such as language modeling, demonstrating promising performance. To foster their reliable use in real-world scenarios, it is crucial to augment their transparency. Our work bridges this critical gap by bringing explainability, particularly Layer-wise Relevance Propagation (LRP), to the Mamba architecture. Guided by the axiom of relevance conservation, we identify specific components in the Mamba architecture, which cause unfaithful explanations. To remedy this issue, we propose MambaLRP, a novel algorithm within the LRP framework, which ensures a more stable and reliable relevance propagation through these components. Our proposed method is theoretically sound and excels in achieving state-of-the-art explanation performance across a diverse range of models and datasets. Moreover, MambaLRP facilitates a deeper inspection of Mamba architectures, uncovering various biases and evaluating their significance. It also enables the analysis of previous speculations regarding the long-range capabilities of Mamba models.

Negation is a common linguistic phenomenon. Yet language models face challenges with negation in many natural language understanding tasks such as question answering and natural language inference. In this paper, we experiment with seamless strategies that incorporate affirmative interpretations (i.e., paraphrases without negation) to make models more robust against negation. Crucially, our affirmative interpretations are obtained automatically. We show improvements with CondaQA, a large corpus requiring reasoning with negation, and five natural language understanding tasks., Comment: Accepted to ACL 2024

Recent text-to-image diffusion-based generative models have the stunning ability to generate highly detailed and photo-realistic images and achieve state-of-the-art low FID scores on challenging image generation benchmarks. However, one of the primary failure modes of these text-to-image generative models is in composing attributes, objects, and their associated relationships accurately into an image. In our paper, we investigate this compositionality-based failure mode and highlight that imperfect text conditioning with CLIP text-encoder is one of the primary reasons behind the inability of these models to generate high-fidelity compositional scenes. In particular, we show that (i) there exists an optimal text-embedding space that can generate highly coherent compositional scenes which shows that the output space of the CLIP text-encoder is sub-optimal, and (ii) we observe that the final token embeddings in CLIP are erroneous as they often include attention contributions from unrelated tokens in compositional prompts. Our main finding shows that the best compositional improvements can be achieved (without harming the model's FID scores) by fine-tuning {\it only} a simple linear projection on CLIP's representation space in Stable-Diffusion variants using a small set of compositional image-text pairs. This result demonstrates that the sub-optimality of the CLIP's output space is a major error source. We also show that re-weighting the erroneous attention contributions in CLIP can also lead to improved compositional performances, however these improvements are often less significant than those achieved by solely learning a linear projection head, highlighting erroneous attentions to be only a minor error source.

Despite the proliferation of wearable health trackers and the importance of sleep and exercise to health, deriving actionable personalized insights from wearable data remains a challenge because doing so requires non-trivial open-ended analysis of these data. The recent rise of large language model (LLM) agents, which can use tools to reason about and interact with the world, presents a promising opportunity to enable such personalized analysis at scale. Yet, the application of LLM agents in analyzing personal health is still largely untapped. In this paper, we introduce the Personal Health Insights Agent (PHIA), an agent system that leverages state-of-the-art code generation and information retrieval tools to analyze and interpret behavioral health data from wearables. We curate two benchmark question-answering datasets of over 4000 health insights questions. Based on 650 hours of human and expert evaluation we find that PHIA can accurately address over 84% of factual numerical questions and more than 83% of crowd-sourced open-ended questions. This work has implications for advancing behavioral health across the population, potentially enabling individuals to interpret their own wearable data, and paving the way for a new era of accessible, personalized wellness regimens that are informed by data-driven insights., Comment: 38 pages

Visual prompting infuses visual information into the input image to adapt models toward specific predictions and tasks. Recently, manually crafted markers such as red circles are shown to guide the model to attend to a target region on the image. However, these markers only work on models trained with data containing those markers. Moreover, finding these prompts requires guesswork or prior knowledge of the domain on which the model is trained. This work circumvents manual design constraints by proposing to learn the visual prompts for guiding the attention of vision transformers. The learned visual prompt, added to any input image would redirect the attention of the pre-trained vision transformer to its spatial location on the image. Specifically, the prompt is learned in a self-supervised manner without requiring annotations and without fine-tuning the vision transformer. Our experiments demonstrate the effectiveness of the proposed optimization-based visual prompting strategy across various pre-trained vision encoders., Comment: Short version (4-pages) accepted as a spotlight paper at T4V workshop, CVPR 2024

Overparametrized transformer networks are the state-of-the-art architecture for Large Language Models (LLMs). However, such models contain billions of parameters making large compute a necessity, while raising environmental concerns. To address these issues, we propose FinerCut, a new form of fine-grained layer pruning, which in contrast to prior work at the transformer block level, considers all self-attention and feed-forward network (FFN) layers within blocks as individual pruning candidates. FinerCut prunes layers whose removal causes minimal alternation to the model's output -- contributing to a new, lean, interpretable, and task-agnostic pruning method. Tested across 9 benchmarks, our approach retains 90% performance of Llama3-8B with 25% layers removed, and 95% performance of Llama3-70B with 30% layers removed, all without fine-tuning or post-pruning reconstruction. Strikingly, we observe intriguing results with FinerCut: 42% (34 out of 80) of the self-attention layers in Llama3-70B can be removed while preserving 99% of its performance -- without additional fine-tuning after removal. Moreover, FinerCut provides a tool to inspect the types and locations of pruned layers, allowing to observe interesting pruning behaviors. For instance, we observe a preference for pruning self-attention layers, often at deeper consecutive decoder layers. We hope our insights inspire future efficient LLM architecture designs., Comment: 22 pages

Spheroids are aggregates of cells that can mimic the cellular organization often found in tissues. They are typically formed through the self-assembly of cells in a culture where there is a promotion of interactions and cell-to-cell communication. Spheroids can be created from various cell types, including cancer cells, stem cells, and primary cells, and they serve as valuable tools in biological research. In this letter, molecule propagation from a point source is simulated in the presence of multiple spheroids to observe the impact of the spheroids on the spatial molecule distribution. The spheroids are modeled as porous media with a corresponding effective diffusion coefficient. System variations are considered with a higher spheroid porosity (i.e., with a higher effective diffusion coefficient) and with molecule uptake by the spheroid cells (approximated as a first-order degradation reaction while molecules diffuse within the spheroid). Results provide initial insights about the molecule propagation dynamics and their potential to model transport and drug delivery within crowded spheroid systems., Comment: submitted to TMBMC journal

The positivistic assumptions of determinism and objectivism in the realm of Newtonian mechanics are questioned in this paper. While objectivism is only challenged through proposing the mildest form of subjectivism, determinism is structurally disputed by proposing that the physical reality, at least in the examples discussed, is, in essence, probabilistic and unpredictable. It is discussed that the physical reality and experimenter's identification of it could basically have inconsistent characteristics, and the study of the physical reality can therefore be conducted in ontic and epistemic levels, leading to two distinct identifications. Four scientific topics, showing two different types of indeterminacy, are introduced and briefly reviewed: chaotic systems, turbulence, fluid transport in porous media, and hydromechanics of fractures. It is proposed that determinism is only meaningful in epistemic level, the first two examples are ontically indeterministic, the last two examples are epistemically indeterministic, and more examples of indeterministic phenomena could, most likely, be found in the nature. Indeterminacy of the physical reality, it is discussed, has always been considered in engineering design processes and such effects have normally been covered through safety factors and feedback loops. By reviewing Hadamard's well-posedness criteria, Poincae's complete deterministic approach, and Leibniz's principles of sufficient reason and identity of indiscernibles, it is claimed that positivism stands on Leibniz's metaphysical assumptions, which are not necessarily in full agreement with the physical reality. A few suggestions for a path beyond positivism in Newtonian mechanics are finally provided.

This paper introduces alternators, a novel family of non-Markovian dynamical models for sequences. An alternator features two neural networks: the observation trajectory network (OTN) and the feature trajectory network (FTN). The OTN and the FTN work in conjunction, alternating between outputting samples in the observation space and some feature space, respectively, over a cycle. The parameters of the OTN and the FTN are not time-dependent and are learned via a minimum cross-entropy criterion over the trajectories. Alternators are versatile. They can be used as dynamical latent-variable generative models or as sequence-to-sequence predictors. When alternators are used as generative models, the FTN produces interpretable low-dimensional latent variables that capture the dynamics governing the observations. When alternators are used as sequence-to-sequence predictors, the FTN learns to predict the observed features. In both cases, the OTN learns to produce sequences that match the data. Alternators can uncover the latent dynamics underlying complex sequential data, accurately forecast and impute missing data, and sample new trajectories. We showcase the capabilities of alternators in three applications. We first used alternators to model the Lorenz equations, often used to describe chaotic behavior. We then applied alternators to Neuroscience, to map brain activity to physical activity. Finally, we applied alternators to Climate Science, focusing on sea-surface temperature forecasting. In all our experiments, we found alternators are stable to train, fast to sample from, yield high-quality generated samples and latent variables, and outperform strong baselines such as neural ODEs and diffusion models in the domains we studied., Comment: A new versatile family of sequence models that can be used for both generative modeling and supervised learning. The codebase will be made available upon publication. This paper is dedicated to Thomas Sankara

We propose a novel finite element-based physics-informed operator learning framework that allows for predicting spatiotemporal dynamics governed by partial differential equations (PDEs). The proposed framework employs a loss function inspired by the finite element method (FEM) with the implicit Euler time integration scheme. A transient thermal conduction problem is considered to benchmark the performance. The proposed operator learning framework takes a temperature field at the current time step as input and predicts a temperature field at the next time step. The Galerkin discretized weak formulation of the heat equation is employed to incorporate physics into the loss function, which is coined finite operator learning (FOL). Upon training, the networks successfully predict the temperature evolution over time for any initial temperature field at high accuracy compared to the FEM solution. The framework is also confirmed to be applicable to a heterogeneous thermal conductivity and arbitrary geometry. The advantages of FOL can be summarized as follows: First, the training is performed in an unsupervised manner, avoiding the need for a large data set prepared from costly simulations or experiments. Instead, random temperature patterns generated by the Gaussian random process and the Fourier series, combined with constant temperature fields, are used as training data to cover possible temperature cases. Second, shape functions and backward difference approximation are exploited for the domain discretization, resulting in a purely algebraic equation. This enhances training efficiency, as one avoids time-consuming automatic differentiation when optimizing weights and biases while accepting possible discretization errors. Finally, thanks to the interpolation power of FEM, any arbitrary geometry can be handled with FOL, which is crucial to addressing various engineering application scenarios.

This work investigates the evolution of proto-neutron stars (PNSs) from birth as neutrino-rich objects to maturity as cold-catalyzed neutrino-poor objects with nucleonic and non-nucleonic degrees of freedom. The focus is on the star's core where the nucleons, hyperons, and the $\Delta$-isobars are expected to dissolve into a "soup" of deconfined quarks, at higher baryon densities, to establish a possible hadron-quark phase transition. We separately calculate the nuclear equations of state (EoS) for the hadronic matter (composed of all the baryon octet and $\Delta$-isobars) and the strange quark matter (SQM) under the same thermodynamic conditions characteristic of PNS and proto-strange star (PSS) evolution and construct the hybrid EoS using Maxwell's construction. The study allows us to determine the hadron-quark phase transitions along the evolution lines of the star. We observed a phase transition from hadronic matter to quark matter (QM) phase when the neutrinos have completely escaped from the star's core. The EoSs utilized are constrained to meet the $2\,\rm M_\odot$ threshold in accordance with the observational data., Comment: 20 pages, 10 figures

Understanding the 3D structure of the Milky Way is a crucial step in deriving properties of the star-forming regions, as well as the Galaxy as a whole. We present a novel 3D map of the Milky Way plane that extends to 10 kpc distance from the Sun. We leverage the wealth of information in the near-IR APOGEE dataset and combine that with our state-of-the-art 3D mapping technique using Bayesian statistics and the Gaussian process to provide a large-scale 3D map of the dust in the Milky Way. Our map stretches across 10 kpc along both the X and Y axes, and 750 pc in the Z direction, perpendicular to the Galactic plane. Our results reveal multi-scale over-densities as well as large cavities in the Galactic plane and shed new light on the Galactic structure and spiral arms. We also provide a catalogue of large molecular clouds identified by our map with accurate distance and volume density estimates. Utilising volume densities derived from this map, we explore mass distribution across various Galactocentric radii. A general decline towards the outer Galaxy is observed, followed by local peaks, some aligning with established features like the Molecular Ring and segments of the spiral arms. Moreover, this work explores extragalactic observational effects on derived properties of molecular clouds by demonstrating the potential biases arising from column density measurements in inferring properties of these regions, and opens exciting avenues for further exploration and analysis, offering a deeper perspective on the complex processes that shape our galaxy and beyond., Comment: Revised version after first referee report; to appear in A&A. Includes SRKh's goodbye letter to Astronomy!

We construct an integrable sigma model with a generalized $\mathcal{F}$ structure, which involves a generalized Nijenhuis structure $\mathcal{J}$ satisfying $\mathcal{J}^{3}=-\mathcal{J}$. Utilizing the expression of the generalized complex structure on the metric Lie group manifold $G$ in terms of operator relations on its Lie algebra $\mathfrak{g}$, we formulate a Yang-Baxter sigma model with a generalized complex structure. Additionally, we present multi-Yang-Baxter sigma models featuring two and three compatible Nijenhuis structures. Examples for each of these models are provided., Comment: 14 pages, two refs are added

Mean-field games (MFG) provide a statistical physics inspired modeling framework for decision making in large-populations of strategic, non-cooperative agents. Mathematically, these systems consist of a forward-backward in time system of two coupled nonlinear partial differential equations (PDEs), namely the Fokker-Plank and the Hamilton-Jacobi-Bellman equations, governing the agent state and control distribution, respectively. In this work, we study a finite-time MFG with a rich global bifurcation structure using a reduced-order model (ROM). The ROM is a 4D two-point boundary value problem obtained by restricting the controlled dynamics to first two moments of the agent state distribution, i.e., the mean and the variance. Phase space analysis of the ROM reveals that the invariant manifolds of periodic orbits around the so-called `ergodic MFG equilibrium' play a crucial role in determining the bifurcation diagram, and impart a topological signature to various solution branches. We show a qualitative agreement of these results with numerical solutions of the full-order MFG PDE system., Comment: 32 pages, 16 figures

In this study, we develop a novel multi-fidelity deep learning approach that transforms low-fidelity solution maps into high-fidelity ones by incorporating parametric space information into a standard autoencoder architecture. This method's integration of parametric space information significantly reduces the need for training data to effectively predict high-fidelity solutions from low-fidelity ones. In this study, we examine a two-dimensional steady-state heat transfer analysis within a highly heterogeneous materials microstructure. The heat conductivity coefficients for two different materials are condensed from a 101 x 101 grid to smaller grids. We then solve the boundary value problem on the coarsest grid using a pre-trained physics-informed neural operator network known as Finite Operator Learning (FOL). The resulting low-fidelity solution is subsequently upscaled back to a 101 x 101 grid using a newly designed enhanced autoencoder. The novelty of the developed enhanced autoencoder lies in the concatenation of heat conductivity maps of different resolutions to the decoder segment in distinct steps. Hence the developed algorithm is named microstructure-embedded autoencoder (MEA). We compare the MEA outcomes with those from finite element methods, the standard U-Net, and various other upscaling techniques, including interpolation functions and feedforward neural networks (FFNN). Our analysis shows that MEA outperforms these methods in terms of computational efficiency and error on test cases. As a result, the MEA serves as a potential supplement to neural operator networks, effectively upscaling low-fidelity solutions to high fidelity while preserving critical details often lost in traditional upscaling methods, particularly at sharp interfaces like those seen with interpolation.

Identifying layers within text-to-image models which control visual attributes can facilitate efficient model editing through closed-form updates. Recent work, leveraging causal tracing show that early Stable-Diffusion variants confine knowledge primarily to the first layer of the CLIP text-encoder, while it diffuses throughout the UNet.Extending this framework, we observe that for recent models (e.g., SD-XL, DeepFloyd), causal tracing fails in pinpointing localized knowledge, highlighting challenges in model editing. To address this issue, we introduce the concept of Mechanistic Localization in text-to-image models, where knowledge about various visual attributes (e.g., "style", "objects", "facts") can be mechanistically localized to a small fraction of layers in the UNet, thus facilitating efficient model editing. We localize knowledge using our method LocoGen which measures the direct effect of intermediate layers to output generation by performing interventions in the cross-attention layers of the UNet. We then employ LocoEdit, a fast closed-form editing method across popular open-source text-to-image models (including the latest SD-XL)and explore the possibilities of neuron-level model editing. Using Mechanistic Localization, our work offers a better view of successes and failures in localization-based text-to-image model editing. Code will be available at https://github.com/samyadeepbasu/LocoGen., Comment: Appearing in ICML 2024

In this paper, we investigate the joint resource allocation and trajectory design for a multi-user, multi-target unmanned aerial vehicle (UAV)-enabled integrated sensing and communication (ISAC) system, where the link capacity between a ground base station (BS) and the UAV is limited. The UAV conducts target sensing and information transmission in orthogonal time slots to prevent interference. As is common in practical systems, sensing is performed while the UAV hovers, allowing the UAV to acquire high-quality sensing data. Subsequently, the acquired sensing data is offloaded to the ground BS for further processing. We jointly optimize the UAV trajectory, UAV velocity, beamforming for the communication users, power allocated to the sensing beam, and time of hovering for sensing to minimize the power consumption of the UAV while ensuring the communication quality of service (QoS) and successful sensing. Due to the prohibitively high complexity of the resulting non-convex mixed integer non-linear program (MINLP), we employ a series of transformations and optimization techniques, including semidefinite relaxation, big-M method, penalty approach, and successive convex approximation, to obtain a low-complexity suboptimal solution. Our simulation results reveal that 1) the proposed design achieves significant power savings compared to two baseline schemes; 2) stricter sensing requirements lead to longer sensing times, highlighting the challenge of efficiently managing both sensing accuracy and sensing time; 3) the optimized trajectory design ensures precise hovering directly above the targets during sensing, enhancing sensing quality and enabling the application of energy-focused beams; and 4) the proposed trajectory design balances the capacity of the backhaul link and the downlink rate of the communication users., Comment: This paper is accepted by IEEE Transactions on Wireless Communications

In this paper, we present a novel extension of massive gravity theory; the Brane-Chern-Simons massive gravity theory. We explore the cosmological implications of this theory by deriving the background equations and demonstrating the existence of self-accelerating solutions. Interestingly, our theory suggests the existence of self-accelerating mechanisms that originate from an effective cosmological constant, leading to intriguing possibilities for understanding the nature of cosmic acceleration. Furthermore, we perform a tensor perturbation analysis to investigate the propagation of gravitational waves in this framework. We derive the dispersion relation for gravitational waves and study their behavior in the Friedmann-Lema\^itre-Roberson-Waker cosmology within the context of Brane-Chen-Simons massive gravity. Utilizing the latest Union2 type Ia supernova dataset comprising $557$ SNIa events, we provide observational support for our theoretical framework, indicating that the Brane-Chern-Simons massive gravity theory is consistent with cosmological observations.

The Standard Model is a theory that accurately predicts the interactions of fundamental particles. However, in recent years, certain phenomena have arisen that cannot be explained by this theory. One particular challenge is the anomalous magnetic moment of the muon. Various theories have been proposed in the Beyond Standard Model, with the aim of addressing these anomalies. One such theory is the Noncommutative Space-Time, which places importance on the noncommutative parameter ($\theta ^{\mu \nu}$). Different values for this parameter have been explored in various articles. However, in this particular article, we utilize the most precise values of $ (240 \, TeV)^{-2}$ for muons and $(3000 \, TeV)^{-2}$ for electrons, in order to align our findings more closely with reality. The tree level diagram for the muon in the Seiberg-Witten map has not yet been calculated. However, we have performed this calculation and predict the anomalous magnetic moment of the muon to be approximately $10^{-10}$. The Noncommutative Space-Time theory yields accurate results in both high energy and low energy physics. Furthermore, the predicted electric dipole moment for the muon and electron, using the parameters outlined in this study are about $10^{-24}$ e.cm for the muon and $10^{-34}$ e.cm for the electron., Comment: 6 pages, 2 figures

The nuclear reduced cross section $\sigma^{A}_{r}$, in the kinematic range of the electron-Ion collider with center-of-mass energy $\sqrt{s}=140~\mathrm{GeV}$ and $y{\leq}1$, is discussed. The importance of the nuclear longitudinal structure function $F^{A}_{L}$ and its behavior owing to the impact parameter for the heavy and light nucleus of Pb-208 and C-12 at $Q^2=5$ and $10~\mathrm{GeV}^2$ is considered. The dependence of the ratios $R^{A}_{F_{L}}$ and $R^{A}_{\sigma}$ on the impact parameter and the expanding point of the gluon density at small $x$ are investigated. The factorized form of parton distributions in nuclei is used in HIJING2.0 model.

The investigation of non-vacuum cosmological backgrounds containing black holes is greatly enhanced by the Kiselev solution. This solution plays a crucial role in understanding the properties of the background and its relationship with the features of the black hole. Consequently, the gravitational memory effects at large distances from the black hole offer a valuable means of obtaining information about the surrounding field parameter N and parameters related to the hair of the hairy Kiselev Black hole. This paper investigates the gravitational memory effects in the context of the Kiselev solution through two distinct approaches. At first, the gravitational memory effect at null infinity is explored by utilizing the Bondi-Sachs formalism by introducing a gravitational wave (GW) pulse to the solution. The resulting Bondi mass is then analyzed to gain further insight. Therefore, the Kiselev solution is being examined to determine the variations in Bondi mass caused by the pulse of GWs. The study of changes in Bondi mass is motivated by the fact that it is dynamic and time-dependent, and it measures mass on an asymptotically null slice or the densities of energy on celestial spheres. In the second approach, the investigation of displacement and velocity memory effects is undertaken in relation to the deviation of two neighboring geodesics and the deviation of their derivative influenced by surrounding field parameter N and the hair of hairy Kiselev black hole. This analysis is conducted within the context of a gravitational wave pulse present in the background of a hairy Kiselev black hole surrounded by a field parameter N., Comment: 18 pages, 30 figures

This work proposes a novel framework for analyzing disease progression using time-aware neural ordinary differential equations (NODE). We introduce a "time-aware head" in a framework trained through self-supervised learning (SSL) to leverage temporal information in latent space for data augmentation. This approach effectively integrates NODEs with SSL, offering significant performance improvements compared to traditional methods that lack explicit temporal integration. We demonstrate the effectiveness of our strategy for diabetic retinopathy progression prediction using the OPHDIAT database. Compared to the baseline, all NODE architectures achieve statistically significant improvements in area under the ROC curve (AUC) and Kappa metrics, highlighting the efficacy of pre-training with SSL-inspired approaches. Additionally, our framework promotes stable training for NODEs, a commonly encountered challenge in time-aware modeling., Comment: Submitted to MICCAI 2024

We report the observation of a coalescing compact binary with component masses $2.5\text{-}4.5~M_\odot$ and $1.2\text{-}2.0~M_\odot$ (all measurements quoted at the 90% credible level). The gravitational-wave signal GW230529_181500 was observed during the fourth observing run of the LIGO-Virgo-KAGRA detector network on 2023 May 29 by the LIGO Livingston Observatory. The primary component of the source has a mass less than $5~M_\odot$ at 99% credibility. We cannot definitively determine from gravitational-wave data alone whether either component of the source is a neutron star or a black hole. However, given existing estimates of the maximum neutron star mass, we find the most probable interpretation of the source to be the coalescence of a neutron star with a black hole that has a mass between the most massive neutron stars and the least massive black holes observed in the Galaxy. We provisionally estimate a merger rate density of $55^{+127}_{-47}~\text{Gpc}^{-3}\,\text{yr}^{-1}$ for compact binary coalescences with properties similar to the source of GW230529_181500; assuming that the source is a neutron star-black hole merger, GW230529_181500-like sources constitute about 60% of the total merger rate inferred for neutron star-black hole coalescences. The discovery of this system implies an increase in the expected rate of neutron star-black hole mergers with electromagnetic counterparts and provides further evidence for compact objects existing within the purported lower mass gap., Comment: 45 pages (10 pages author list, 13 pages main text, 1 page acknowledgements, 13 pages appendices, 8 pages bibliography), 17 figures, 16 tables. Update to match version published in The Astrophysical Journal Letters. Data products available from https://zenodo.org/records/10845779

Graph-structured data is prevalent in domains such as social networks, financial transactions, brain networks, and protein interactions. As a result, the research community has produced new databases and analytics engines to process such data. Unfortunately, there is not yet widespread benchmark standardization in graph processing, and the heterogeneity of evaluations found in the literature can lead researchers astray. Evaluations frequently ignore datasets' statistical idiosyncrasies, which significantly affect system performance. Scalability studies often use datasets that fit easily in memory on a modest desktop. Some studies rely on synthetic graph generators, but these generators produce graphs with unnatural characteristics that also affect performance, producing misleading results. Currently, the community has no consistent and principled manner with which to compare systems and provide guidance to developers who wish to select the system most suited to their application. We provide three different systematizations of benchmarking practices. First, we present a 12-year literary review of graph processing benchmarking, including a summary of the prevalence of specific datasets and benchmarks used in these papers. Second, we demonstrate the impact of two statistical properties of datasets that drastically affect benchmark performance. We show how different assignments of IDs to vertices, called vertex orderings, dramatically alter benchmark performance due to the caching behavior they induce. We also show the impact of zero-degree vertices on the runtime of benchmarks such as breadth-first search and single-source shortest path. We show that these issues can cause performance to change by as much as 38% on several popular graph processing systems. Finally, we suggest best practices to account for these issues when evaluating graph systems.

The run-time for optimization tools used in chip design has grown with the complexity of designs to the point where it can take several days to go through one design cycle which has become a bottleneck. Designers want fast tools that can quickly give feedback on a design. Using the input and output data of the tools from past designs, one can attempt to build a machine learning model that predicts the outcome of a design in significantly shorter time than running the tool. The accuracy of such models is affected by the representation of the design data, which is usually a netlist that describes the elements of the digital circuit and how they are connected. Graph representations for the netlist together with graph neural networks have been investigated for such models. However, the characteristics of netlists pose several challenges for existing graph learning frameworks, due to the large number of nodes and the importance of long-range interactions between nodes. To address these challenges, we represent the netlist as a directed hypergraph and propose a Directional Equivariant Hypergraph Neural Network (DE-HNN) for the effective learning of (directed) hypergraphs. Theoretically, we show that our DE-HNN can universally approximate any node or hyperedge based function that satisfies certain permutation equivariant and invariant properties natural for directed hypergraphs. We compare the proposed DE-HNN with several State-of-the-art (SOTA) machine learning models for (hyper)graphs and netlists, and show that the DE-HNN significantly outperforms them in predicting the outcome of optimized place-and-route tools directly from the input netlists. Our source code and the netlists data used are publicly available at https://github.com/YusuLab/chips.git

To obtain fast solutions for governing physical equations in solid mechanics, we introduce a method that integrates the core ideas of the finite element method with physics-informed neural networks and concept of neural operators. This approach generalizes and enhances each method, learning the parametric solution for mechanical problems without relying on data from other resources (e.g. other numerical solvers). We propose directly utilizing the available discretized weak form in finite element packages to construct the loss functions algebraically, thereby demonstrating the ability to find solutions even in the presence of sharp discontinuities. Our focus is on micromechanics as an example, where knowledge of deformation and stress fields for a given heterogeneous microstructure is crucial for further design applications. The primary parameter under investigation is the Young's modulus distribution within the heterogeneous solid system. Our investigations reveal that physics-based training yields higher accuracy compared to purely data-driven approaches for unseen microstructures. Additionally, we offer two methods to directly improve the process of obtaining high-resolution solutions, avoiding the need to use basic interpolation techniques. First is based on an autoencoder approach to enhance the efficiency for calculation on high resolution grid point. Next, Fourier-based parametrization is utilized to address complex 2D and 3D problems in micromechanics. The latter idea aims to represent complex microstructures efficiently using Fourier coefficients. Comparisons with other well-known operator learning algorithms, further emphasize the advantages of the newly proposed method.

In this paper we have performed a comparative study of different types of oscillating dark energy models using the Metropolis algorithm of MCMC. Eight different oscillating parameterization being examined herein which have demonstrated considerable ability to fit the overall cosmological observations including Pantheon sample of SnIa, Baryon Acoustic Oscillations, Cosmic Chronometer Hubble data and distance priors of Planck CMB. In order to compare the consistency of these models with observations, we have used both Akaike and Deviation information criteria. Although, the values of Akaike information criteria for different models indicate that there is no support for oscillating DE models, Deviation information criteria showed that there is significant support for some of these models. Our results showed that these models are capable of solving cosmic coincidence problem and alleviating the Hubble tension. Comparing the $H_0$ values obtained for different oscillating scenarios, with that of $\Lambda$CDM, we observe that our oscillating models led to $\bar{H_0}=69.78$ which is $0.29$ greater than $H_{0,\Lambda}$ and thus reduce the Hubble tension. Among all of models, Model(1) with $H_0=70.00 \pm 0.71$ is the most capable of alleviating the $H_0$ tension. Furthermore, we examined our models assuming $H_0=73.0 \pm 1.4$ from SHoES measurements. We find that adding this data point to our data combination, led to a $\Delta \bar{H_0}=0.95$ increase in $H_0$ value for different models., Comment: 11 pages, 2 tables and 4 figures, Submitted in APJ

Pre-training strategies based on self-supervised learning (SSL) have proven to be effective pretext tasks for many downstream tasks in computer vision. Due to the significant disparity between medical and natural images, the application of typical SSL is not straightforward in medical imaging. Additionally, those pretext tasks often lack context, which is critical for computer-aided clinical decision support. In this paper, we developed a longitudinal masked auto-encoder (MAE) based on the well-known Transformer-based MAE. In particular, we explored the importance of time-aware position embedding as well as disease progression-aware masking. Taking into account the time between examinations instead of just scheduling them offers the benefit of capturing temporal changes and trends. The masking strategy, for its part, evolves during follow-up to better capture pathological changes, ensuring a more accurate assessment of disease progression. Using OPHDIAT, a large follow-up screening dataset targeting diabetic retinopathy (DR), we evaluated the pre-trained weights on a longitudinal task, which is to predict the severity label of the next visit within 3 years based on the past time series examinations. Our results demonstrated the relevancy of both time-aware position embedding and masking strategies based on disease progression knowledge. Compared to popular baseline models and standard longitudinal Transformers, these simple yet effective extensions significantly enhance the predictive ability of deep classification models.