Search

DNA-Encoded Libraries (DEL) are combinatorial small molecule libraries that offer an efficient way to characterize diverse chemical spaces. Selection experiments using DELs are pivotal to drug discovery efforts, enabling high-throughput screens for hit finding. However, limited availability of public DEL datasets hinders the advancement of computational techniques designed to process such data. To bridge this gap, we present KinDEL, one of the first large, publicly available DEL datasets on two kinases: Mitogen-Activated Protein Kinase 14 (MAPK14) and Discoidin Domain Receptor Tyrosine Kinase 1 (DDR1). Interest in this data modality is growing due to its ability to generate extensive supervised chemical data that densely samples around select molecular structures. Demonstrating one such application of the data, we benchmark different machine learning techniques to develop predictive models for hit identification; in particular, we highlight recent structure-based probabilistic approaches. Finally, we provide biophysical assay data, both on- and off-DNA, to validate our models on a smaller subset of molecules. Data and code for our benchmarks can be found at: https://github.com/insitro/kindel.

Recent advances in Deep Neural Networks (DNNs) have demonstrated outstanding performance across various domains. However, their large size is a challenge for deployment on resource-constrained devices such as mobile, edge, and IoT platforms. To overcome this, a distributed inference setup can be used where a small-sized DNN (initial few layers) can be deployed on mobile, a bigger version on the edge, and the full-fledged, on the cloud. A sample that has low complexity (easy) could be then inferred on mobile, that has moderate complexity (medium) on edge, and higher complexity (hard) on the cloud. As the complexity of each sample is not known beforehand, the following question arises in distributed inference: how to decide complexity so that it is processed by enough layers of DNNs. We develop a novel approach named DIMEE that utilizes Early Exit (EE) strategies developed to minimize inference latency in DNNs. DIMEE aims to improve the accuracy, taking into account the offloading cost from mobile to edge/cloud. Experimental validation on GLUE datasets, encompassing various NLP tasks, shows that our method significantly reduces the inference cost (> 43%) while maintaining a minimal drop in accuracy (< 0.3%) compared to the case where all the inference is made in cloud., Comment: 8 pages, 3 figures

Deep neural networks (DNNs) have made significant progress in recognizing visual elements and generating descriptive text in image-captioning tasks. However, their improved performance comes from increased computational burden and inference latency. Early Exit (EE) strategies can be used to enhance their efficiency, but their adaptation presents challenges in image captioning as it requires varying levels of semantic information for accurate predictions. To overcome this, we introduce CAPEEN to improve the performance of EE strategies using knowledge distillation. Inference in CAPEEN is completed at intermediary layers if prediction confidence exceeds a predefined value learned from the training data. To account for real-world deployments, where target distributions could drift from that of training samples, we introduce a variant A-CAPEEN to adapt the thresholds on the fly using Multiarmed bandits framework. Experiments on the MS COCO and Flickr30k datasets show that CAPEEN gains speedup of 1.77x while maintaining competitive performance compared to the final layer, and A-CAPEEN additionally offers robustness against distortions. The source code is available at https://github.com/Div290/CapEEN, Comment: To appear in EMNLP (finding) 2024

Pre-trained Language Models (PLMs) exhibit good accuracy and generalization ability across various tasks using self-supervision, but their large size results in high inference latency. Early Exit (EE) strategies handle the issue by allowing the samples to exit from classifiers attached to the intermediary layers, but they do not generalize well, as exit classifiers can be sensitive to domain changes. To address this, we propose Unsupervised Domain Adaptation in EE framework (DADEE) that employs multi-level adaptation using knowledge distillation. DADEE utilizes GAN-based adversarial adaptation at each layer to achieve domain-invariant representations, reducing the domain gap between the source and target domain across all layers. The attached exits not only speed up inference but also enhance domain adaptation by reducing catastrophic forgetting and mode collapse, making it more suitable for real-world scenarios. Experiments on tasks such as sentiment analysis, entailment classification, and natural language inference demonstrate that DADEE consistently outperforms not only early exit methods but also various domain adaptation methods under domain shift scenarios. The anonymized source code is available at https://github.com/Div290/DAdEE., Comment: To appear in EMNLP (findings) 2024

Cononsolvency occurs when two miscible, competing good solvents for a polymer are mixed, resulting in a loss of solubility. In this study, we demonstrate through simulations, supported by theory, that cononsolvency can be driven solely by solvent-cosolvent attraction ($\epsilon_{sc}$). The primary mechanism underlying this behavior is the emergent depletion effect, which is amplified by solvent-cosolvent interactions. The polymer reaches a compact state when the solvent and cosolvent fractions are equal ($x_s = x_c = 0.5$), a finding that aligns with predictions from Flory-Huggins theory and the random phase approximation. We show that this cononsolvency behavior is observed for different cosolvent sizes, provided the cosolvent density remains below the depletion threshold and the sizes of solvent and cosolvent particles are not smaller than the monomer size. Additionally, we investigate the role of temperature and find that cononsolvency weakens as temperature increases, due to a reduction in the depletion effect. Finally, we show that when preferential cosolvent attraction is introduced in this simple model, it leads to cononsolvency driven by bridging interactions, occurring at lower cosolvent fractions ($x_c < 0.5$)., Comment: 10 pages and an ancillary file

This paper introduces an integer-valued generalized autoregressive conditional heteroskedasticity (INGARCH) model based on the novel geometric distribution and discusses some of its properties. The parameter estimation problem of the models are studied by conditional maximum likelihood and Bayesian approach using Hamiltonian Monte Carlo (HMC) algorithm. The results of the simulation studies and real data analysis affirm the good performance of the estimators and the model.

Large Language Models (LLMs) have succeeded considerably in In-Context-Learning (ICL) based summarization. However, saliency is subject to the users' specific preference histories. Hence, we need reliable In-Context Personalization Learning (ICPL) capabilities within such LLMs. For any arbitrary LLM to exhibit ICPL, it needs to have the ability to discern contrast in user profiles. A recent study proposed a measure for degree-of-personalization called EGISES for the first time. EGISES measures a model's responsiveness to user profile differences. However, it cannot test if a model utilizes all three types of cues provided in ICPL prompts: (i) example summaries, (ii) user's reading histories, and (iii) contrast in user profiles. To address this, we propose the iCOPERNICUS framework, a novel In-COntext PERsonalization learNIng sCrUtiny of Summarization capability in LLMs that uses EGISES as a comparative measure. As a case-study, we evaluate 17 state-of-the-art LLMs based on their reported ICL performances and observe that 15 models' ICPL degrades (min: 1.6%; max: 3.6%) when probed with richer prompts, thereby showing lack of true ICPL.

We theoretically investigated an array of coupled optomechanical cavities driven by a phase-modulated laser. We show that phase modulation enables the control of band structure and switching of the relative weights of photons and phonons in hybrid eigenmodes. Finally, we show how phase affects the population of hybrid modes and quench dynamics., Comment: 10 pages, 7 figures

The anisotropic light-matter interactions in 2D materials have garnered significant attention for their potential to develop futuristic polarization-based optoelectronic devices, such as photodetectors and photo-actuators. In this study, we investigate the polarization-dependent interactions in ferroelectric 3R alpha-In2Se3 using Angle-Resolved Polarized Raman Spectroscopy (ARPRS) with different excitation lasers. Our experimental findings supported by complementary Density Functional Theory calculations demonstrate that the light-matter interactions depend not only on the crystallographic orientation but also on the excitation energy. Scanning transmission electron microscopy (STEM) confirms the highly anisotropic 3R crystal structure of alpha-In2Se3. This anisotropy in crystal structure facilitates significant optical anisotropy, driven by a complex interplay of electron-photon and electron-phonon interactions, which is reflected in the complex nature of the Raman tensor elements. These anisotropy interactions extend to the materials electrical response under light illumination. Remarkably, the anisotropic photo-response can be tuned by both polarization and wavelength of the incident light, making In2Se3 a promising material for advanced polarization-sensitive photodetection applications., Comment: 19 pages, 6 figures, supporting information

We present ImPoster, a novel algorithm for generating a target image of a 'source' subject performing a 'driving' action. The inputs to our algorithm are a single pair of a source image with the subject that we wish to edit and a driving image with a subject of an arbitrary class performing the driving action, along with the text descriptions of the two images. Our approach is completely unsupervised and does not require any access to additional annotations like keypoints or pose. Our approach builds on a pretrained text-to-image latent diffusion model and learns the characteristics of the source and the driving image by finetuning the diffusion model for a small number of iterations. At inference time, ImPoster performs step-wise text prompting i.e. it denoises by first moving in the direction of the image manifold corresponding to the driving image followed by the direction of the image manifold corresponding to the text description of the desired target image. We propose a novel diffusion guidance formulation, image frequency guidance, to steer the generation towards the manifold of the source subject and the driving action at every step of the inference denoising. Our frequency guidance formulations are derived from the frequency domain properties of images. We extensively evaluate ImPoster on a diverse set of source-driving image pairs to demonstrate improvements over baselines. To the best of our knowledge, ImPoster is the first approach towards achieving both subject-driven as well as action-driven image personalization. Code and data is available at https://github.com/divyakraman/ImPosterDiffusion2024.

Inductive spatial temporal prediction can generalize historical data to predict unseen data, crucial for highly dynamic scenarios (e.g., traffic systems, stock markets). However, external events (e.g., urban structural growth, market crash) and emerging new entities (e.g., locations, stocks) can undermine prediction accuracy by inducing data drift over time. Most existing studies extract invariant patterns to counter data drift but ignore pattern diversity, exhibiting poor generalization to unseen entities. To address this issue, we design an Informative Graph Neural Network (INF-GNN) to distill diversified invariant patterns and improve prediction accuracy under data drift. Firstly, we build an informative subgraph with a uniquely designed metric, Relation Importance (RI), that can effectively select stable entities and distinct spatial relationships. This subgraph further generalizes new entities' data via neighbors merging. Secondly, we propose an informative temporal memory buffer to help the model emphasize valuable timestamps extracted using influence functions within time intervals. This memory buffer allows INF-GNN to discern influential temporal patterns. Finally, RI loss optimization is designed for pattern consolidation. Extensive experiments on real-world dataset under substantial data drift demonstrate that INF-GNN significantly outperforms existing alternatives.

We study approximately orthogonality (in the sense of Dragomir) preserving and reversing operators. We obtain a complete characterization of approximate orthogonality preserving and reversing operators for a class of operators. We also study the locally approximate orthogonality preserving and reversing operators defined on some finite-dimensional Banach spaces.

We introduce the notion of local orthogonality preserving operators to study the right-symmetry of operators. As a consequence of our work, we show that any smooth compact operator defined on a smooth and reflexive Banach space is either a rank one operator or it is not right-symmetric. We show that there are no right-symmetric smooth compact operators defined on a smooth and reflexive Banach space that fails to have any non-zero left-symmetric point. We also study approximately orthogonality preserving and reversing operators (in the sense of Chmieli\'{n}ski and Dragomir). We show that on a finite-dimensional Banach space, an operator is approximately orthogonality preserving (reversing) in the sense of Dragomir if and only if it is an injective operator.

Recommender systems, while a powerful decision making tool, are often operationalized as black box models, such that their AI algorithms are not accessible or interpretable by human operators. This in turn can cause confusion and frustration for the operator and result in unsatisfactory outcomes. While the field of explainable AI has made remarkable strides in addressing this challenge by focusing on interpreting and explaining the algorithms to human operators, there are remaining gaps in the human's understanding of the recommender system. This paper investigates the relative impact of using context, properties of the decision making task and environment, to align human and AI algorithm understanding of the state of the world, i.e. judgment, to improve joint human-recommender performance as compared to utilizing post-hoc algorithmic explanations. We conducted an empirical, between-subjects experiment in which participants were asked to work with an automated recommender system to complete a decision making task. We manipulated the method of transparency (shared contextual information to support shared judgment vs algorithmic explanations) and record the human's understanding of the task, the recommender system, and their overall performance. We found that both techniques yielded equivalent agreement on final decisions. However, those who saw task context had less tendency to over-rely on the recommender system and were able to better pinpoint in what conditions the AI erred. Both methods improved participants' confidence in their own decision making, and increased mental demand equally and frustration negligibly. These results present an alternative approach to improving team performance to post-hoc explanations and illustrate the impact of judgment on human cognition in working with recommender systems., Comment: arXiv admin note: text overlap with arXiv:2310.11370

The attention mechanism is a key computing kernel of Transformers, calculating pairwise correlations across the entire input sequence. The computing complexity and frequent memory access in computing self-attention put a huge burden on the system especially when the sequence length increases. This paper presents an analog and digital hybrid processor to accelerate the attention mechanism for transformers in 65nm CMOS technology. We propose an analog computing-in-memory (CIM) core, which prunes ~75% of low-score tokens on average during runtime at ultra-low power and delay. Additionally, a digital processor performs precise computations only for ~25% unpruned tokens selected by the analog CIM core, preventing accuracy degradation. Measured results show peak energy efficiency of 14.8 and 1.65 TOPS/W, and peak area efficiency of 976.6 and 79.4 GOPS/mm$^\mathrm{2}$ in the analog core and the system-on-chip (SoC), respectively., Comment: 4 pages, 9 figures, to be published at ESSERC 2024

Cardiac arrest remains a leading cause of death worldwide, necessitating proactive measures for early detection and intervention. This project aims to develop and assess predictive models for the timely identification of cardiac arrest incidents, utilizing a comprehensive dataset of clinical parameters and patient histories. Employing machine learning (ML) algorithms like XGBoost, Gradient Boosting, and Naive Bayes, alongside a deep learning (DL) approach with Recurrent Neural Networks (RNNs), we aim to enhance early detection capabilities. Rigorous experimentation and validation revealed the superior performance of the RNN model, which effectively captures complex temporal dependencies within the data. Our findings highlight the efficacy of these models in accurately predicting cardiac arrest likelihood, emphasizing the potential for improved patient care through early risk stratification and personalized interventions. By leveraging advanced analytics, healthcare providers can proactively mitigate cardiac arrest risk, optimize resource allocation, and improve patient outcomes. This research highlights the transformative potential of machine learning and deep learning techniques in managing cardiovascular risk and advances the field of predictive healthcare analytics.

We present the first comprehensive source catalog (UVIT DR1) of ultraviolet (UV) photometry in four far-UV (FUV $\sim$1300$-$1800 \AA) and five near-UV (NUV $\sim$2000$-$3000 \AA) filters of the Ultraviolet Imaging Telescope (UVIT) on board {\em AstroSat}. UVIT DR1 includes bright UV sources in 291 fields that UVIT detected during its first two years of pointed observation, encompassing an area of 58 square degrees. We used the {\sc ccdlab} pipeline to reduce the L1 data, source-extractor for source detection, and four photometric procedures to determine the magnitudes of the detected sources. We provided the 3$\sigma$ and 5$\sigma$ detection limits for all the filters of UVIT. We describe the details of observation, source extraction methods, and photometry procedures applied to prepare the catalog. In the final UVIT DR1 catalog, we have point sources, extended sources, clumps from nearby galaxies, There are 239,520 unique sources in the combined UVIT DR1, of which 70,488 sources have FUV magnitudes, and 211,410 have NUV magnitudes. We cross-matched and compared non-crowded sources of UVIT with the {\em Galaxy Evolution Explorer (GALEX)} and {\em Gaia} source catalogs. We provide a clean catalog of the unique sources in various UVIT filters that will help further multi-wavelength scientific analysis of the objects., Comment: 23 pages, 10 figures, 10 tables, accepted for publication in the Astrophysical Journal Supplement Series

Advanced AI systems capable of generating humanlike text and multimodal content are now widely available. In this paper, we discuss the impacts that generative artificial intelligence may have on democratic processes. We consider the consequences of AI for citizens' ability to make informed choices about political representatives and issues (epistemic impacts). We ask how AI might be used to destabilise or support democratic mechanisms like elections (material impacts). Finally, we discuss whether AI will strengthen or weaken democratic principles (foundational impacts). It is widely acknowledged that new AI systems could pose significant challenges for democracy. However, it has also been argued that generative AI offers new opportunities to educate and learn from citizens, strengthen public discourse, help people find common ground, and to reimagine how democracies might work better., Comment: 25 pages

In the context of multi-domain network slices, multiple domains need to work together to provide a service. The problem of determining which part of the service fits within which domain is referred to as slice partitioning. The partitioning of multi-domain network slices poses a challenging problem, particularly when striving to strike the right balance between inter-domain and intra-domain costs, as well as ensuring optimal load distribution within each domain. To approach the optimal partition solution while maintaining load balance between domains, a framework has been proposed. This framework not only generates partition plans with various characteristics but also employs a Graph Neural Network solver, which significantly reduces the plan generation time. The proposed approach is promising in generating partition plans for multi-domain network slices and is expected to improve the overall performance of the network.

In the 5G era and beyond, it is favorable to deploy latency-sensitive and reliability-aware services on edge computing networks in which the computing and network resources are more limited compared to cloud and core networks but can respond more promptly. These services can be composed as Service Function Chains (SFCs) which consist of a sequence of ordered Virtual Network Functions (VNFs). To achieve efficient edge resources allocation for SFC requests and optimal profit for edge service providers, we formulate the SFC placement problem in an edge environment and propose a multi-agent Reinforcement Learning (RL) scheme to address the problem. The proposed scheme employs a set of RL agents to collaboratively make SFC placement decisions, such as path selection, VNF configuration, and VNF deployment. Simulation results show our model can improve the profit of edge service providers by 12\% compared with a heuristic solution.

Hyperbolic systems of the first and higher-order partial differential equations appear in many multiphysics problems. We will be dealing with a wave propagation problem in a piece-wise homogeneous medium. Mathematically, the problem is reduced to analyzing two systems of partial differential equations posed on two domains with a common boundary. The differential equations may not be satisfied on the boundary (or part of the boundary), but some interface conditions are satisfied. These interface conditions depend on a specific physical problem. We aim to prove the existence and regularity of the solution for the case of hyperbolic systems of first-order equations with different domains separated by a hyperplane, where we need to formulate the interface conditions. We do this for the initial value problem in 1D-space variable when the coefficient matrix has discontinuity on $m$ lines. More specifically, we find explicit solutions to the case when the coefficient matrix is piecewise constant with a discontinuity along $1$ line or $2$ lines. We also prove the existence of solution to the general initial value problem. We then formulate the weak solution of initial value problem for the corresponding symmetric hyperbolic system in $n $D-space variables with interface conditions, get the energy estimates for this system, and prove the existence of solution to the system.

Type-I solar noise storms are perhaps the most commonly observed active radio emissions from the Sun at meter-wavelengths. Noise storms have a long-lived and wideband continuum background with superposed islands of much brighter narrowband and short-lived emissions, known as type-I bursts. There is a serious paucity of studies focusing on the morphology of these two types of emissions, primarily because of the belief that coronal scattering will always wash out any features at small angular scales. However, it is important to { investigate} their spatial structures in detail to make a spatio-temporal connection with observations at extreme-ultraviolet/ X-ray bands to understand the detailed nature of these emissions. In this work, we use high angular resolution observations from the upgraded Giant Metrewave Radio Telescope to demonstrate that it is possible to detect structures with angular scales as small as $\sim 9\arcsec$, about three times smaller than the smallest structure reported to date from noise storms. Our observations also suggest while the individual type-I bursts are narrowband in nature, the bursts are probably caused by traveling disturbance(s) inducing magnetic reconnections at different coronal heights, and thus leading to correlated change in the morphology of the type-I bursts observed at a wide range of frequencies., Comment: Accepted for publication in the Astrophysical Journal

Our experiments on a sphere falling under gravity in Stokes flow show significant history effects. We observe an algebraic, not exponential, relaxation rate to the terminal velocity, validating the solution to the Basset-Boussinesq-Oseen equation. Unlike in steady Stokes theory, our experiments and theory reveal a vortex ring forming around the sphere and drifting away. As the Reynolds number nears unity, the vortex ring lags behind the sphere, departing from Stokesian theory, though the sphere's algebraic response persists. These findings are critical for interactions in the Stokes limit.

The median $M$ of a graph $G$ is the set of vertices with a minimum total distance to all other vertices in the graph. In this paper, we determine the median of Sierpi\'{n}ski triangle graphs. Sierpi\'{n}ski triangle graphs, also known as Sierpi\'{n}ski gasket graphs of order $n$ are graphs formed by contracting all non-clique edges from the Sierpi\'{n}ski graphs of order ($n+1$).

In this work, we explore the effect of neutrino nonstandard interactions (NSI) involving the charm quark at SND@LHC. Using an effective description of new physics in terms of four-fermion operators involving a charm quark, we constrain the Wilson coefficients of the effective interaction from two and three-body charmed meson decays. In our fit, we include charmed meson decays not only to pseudoscalar final states but also to vector final states and include decays to the $\eta$ and $\eta^\prime$ final states. We also consider constraints from charmed baryon decays. We then study the effect of new physics in neutrino scattering processes, involving charm production at SND@LHC, for various benchmark new physics couplings obtained from the low energy fits. Finally, we also study the effects of lepton universality violation (LUV) assuming that the new physics coupling is not lepton universal., Comment: 23 pages, 3 figures; included neutrino production modes that affect results, typos corrected, references added

In continual learning (CL), model growth enhances adaptability over new data, improving knowledge retention for more tasks. However, improper model growth can lead to severe degradation of previously learned knowledge, an issue we name as growth-induced forgetting (GIFt), especially in task-agnostic CL using entire grown model for inference. Existing works, despite adopting model growth and random initialization for better adaptability, often fail to recognize the presence of GIFt caused by improper model growth. This oversight limits comprehensive control of forgetting and hinders full utilization of model growth. We are the first in CL to identify this issue and conduct an in-depth study on root cause of GIFt, where layer expansion stands out among model growth strategies, widening layers without affecting model functionality. Yet, direct adoption of layer expansion presents challenges. It lacks data-driven control and initialization of expanded parameters to balance adaptability and knowledge retention. This paper presents a novel SparseGrow approach to overcome the issue of GIFt while enhancing adaptability over new data. SparseGrow employs data-driven sparse layer expansion to control efficient parameter usage during growth, reducing GIFt from excessive growth and functionality changes. It also combines sparse growth with on-data initialization at training late-stage to create partially 0-valued expansions that fit learned distribution, enhancing retention and adaptability. To further minimize forgetting, freezing is applied by calculating the sparse mask, allowing data-driven preservation of important parameters. Through experiments across datasets with various settings, cases, and task numbers, we demonstrate the necessity of layer expansion and showcase the effectiveness of SparseGrow in overcoming GIFt, highlighting its adaptability and knowledge retention for incremental tasks.

Systems of the first order partial differential equations with singular solutions appear in many multiphysics problems and the weak formulation of solutions involve in many cases product of distributions. In this paper we study such a system derived from Eulerian droplet model for air particle flow. This is a 2 x 2 non - strictly hyperbolic system of conservation laws with linear damping. We first study a regularized viscous system with variable viscosity term and obtain a weak asymptotic solution with general initial data and also get solution in the Colombeau algebra. We also study the vanishing viscosity limit and show that this limit is a distribution solution. Further we study the large time asymptotic behaviour of the viscous system. This important system, is not very well studied due to complexities in the analysis. As far as we know the only work done on this system is for Riemann type of initial data. The significance of this paper is that we work on the system having general initial data and not just initial data of the Riemann type., Comment: Preprint

In this paper we study a system of equations which appear in the modelling of many physical phenomena. Initially this system appeared in description of the large scale structure formation. Recently it is derived as a second order queueing model. We construct weakly asymptotic solutions of the initial boundary value problem for the system and interaction of waves in the quarter plane $\{(x,t): x>0,t>0\}$ with boundary Riemann solution centered at $(0,0)$ and Riemann solution centered at a point $(x_0,0), x_0>0$.

This paper is concerned with the study of interaction of waves originating from the Riemann problem centred at two different points for a system of equations modelling propagation of elastic waves. The system consists of two equations for $(u,\sigma)$, where, u is the velocity and $\sigma$ is the stress and is strictly hyperbolic and nonconservative. Study of interaction of waves is one of the most important steps in the construction of global solution with initial data in the space of functions of bounded variation using approximation procedure like the Glimm's scheme. This amounts to constructing a solution with initial data consisting of three states $(u_L, \sigma_L), (u_m, \sigma_m),$ and $(u_R, \sigma_R)$. Usually this analysis is done for the states which are in a small neighbourhood of a fixed state. Here we get explicit formula for the solution of the system when the data lies in the level sets of Riemann invariants. The speciality of the work is that we donot assume smallness conditions on the initial data.

This paper is a continuation of our previous paper \cite{d1} on the initial boundary value problem for a nonconservative system appearing in elastodynamics in the space time domain $x>0,t>0$. There, the initial and boundary data were assumed to lie on the level sets of one of the Riemann invariants where as in this paper we consider the general initial and boundary data of Riemann type, formulate the boundary value problem based on the Riemann problem and construct explicitly the solution.

Spatial audio in Extended Reality (XR) provides users with better awareness of where virtual elements are placed, and efficiently guides them to events such as notifications, system alerts from different windows, or approaching avatars. Humans, however, are inaccurate in localizing sound cues, especially with multiple sources due to limitations in human auditory perception such as angular discrimination error and front-back confusion. This decreases the efficiency of XR interfaces because users misidentify from which XR element a sound is coming. To address this, we propose Auptimize, a novel computational approach for placing XR sound sources, which mitigates such localization errors by utilizing the ventriloquist effect. Auptimize disentangles the sound source locations from the visual elements and relocates the sound sources to optimal positions for unambiguous identification of sound cues, avoiding errors due to inter-source proximity and front-back confusion. Our evaluation shows that Auptimize decreases spatial audio-based source identification errors compared to playing sound cues at the paired visual-sound locations. We demonstrate the applicability of Auptimize for diverse spatial audio-based interactive XR scenarios., Comment: UIST 2024

Shared caches are vulnerable to side channel attacks through contention in cache sets. Besides being a simple source of information leak, these side channels form useful gadgets for more sophisticated attacks that compromise the security of shared systems. The fundamental design aspect that contention attacks exploit is the deterministic nature of the set of addresses contending for a cache set. In this paper, we present RollingCache, a cache design that defends against contention attacks by dynamically changing the set of addresses contending for cache sets. Unlike prior defenses, RollingCache does not rely on address encryption/decryption, data relocation, or cache partitioning. We use one level of indirection to implement dynamic mapping controlled by the whole-cache runtime behavior. Our solution does not depend on having defined security domains, and can defend against an attacker running on the same or another core. We evaluate RollingCache on ChampSim using the SPEC-2017 benchmark suite. Our security evaluation shows that our dynamic mapping removes the deterministic ability to identify the source of contention. The performance evaluation shows an impact of 1.67\% over a mix of workloads, with a corresponding

This paper is concerned with the initial boundary value problem for a nonconservative system of hyperbolic equation appearing in elastodynamics in the space time domain $x > 0, t > 0$. The number of boundary conditions to be prescribed at the boundary $x = 0$, depend on the number of characteristics entering the domain. Since our system is nonlinear the characteristic speeds depends on the unknown and the direction of the characteristics curves are known apriori. As it is well known, the boundary condition has to be understood in a generalised way. One of the standard way is using vanishing viscosity method. We use this method to construct solution for a particular class of initial and boundary data, namely the initial and boundary datas that lie on the level sets of one of the Riemann invariants., Comment: 13 pages

Anonymity is an important principle online. However, malicious actors have long used misleading identities to conduct fraud, spread disinformation, and carry out other deceptive schemes. With the advent of increasingly capable AI, bad actors can amplify the potential scale and effectiveness of their operations, intensifying the challenge of balancing anonymity and trustworthiness online. In this paper, we analyze the value of a new tool to address this challenge: "personhood credentials" (PHCs), digital credentials that empower users to demonstrate that they are real people -- not AIs -- to online services, without disclosing any personal information. Such credentials can be issued by a range of trusted institutions -- governments or otherwise. A PHC system, according to our definition, could be local or global, and does not need to be biometrics-based. Two trends in AI contribute to the urgency of the challenge: AI's increasing indistinguishability from people online (i.e., lifelike content and avatars, agentic activity), and AI's increasing scalability (i.e., cost-effectiveness, accessibility). Drawing on a long history of research into anonymous credentials and "proof-of-personhood" systems, personhood credentials give people a way to signal their trustworthiness on online platforms, and offer service providers new tools for reducing misuse by bad actors. In contrast, existing countermeasures to automated deception -- such as CAPTCHAs -- are inadequate against sophisticated AI, while stringent identity verification solutions are insufficiently private for many use-cases. After surveying the benefits of personhood credentials, we also examine deployment risks and design challenges. We conclude with actionable next steps for policymakers, technologists, and standards bodies to consider in consultation with the public., Comment: 63 pages, 7 figures, 5 tables; minor additions to acknowledgments and wording changes for clarity; corrected typo

Recent 3D novel view synthesis (NVS) methods are limited to single-object-centric scenes and struggle with complex environments. They often require extensive 3D data for training, lacking generalization beyond the training distribution. Conversely, 3D-free methods can generate text-controlled views of complex, in-the-wild scenes using a pretrained stable diffusion model without the need for a large amount of 3D-based training data, but lack camera control. In this paper, we introduce a method capable of generating camera-controlled viewpoints from a single input image, by combining the benefits of 3D-free and 3D-based approaches. Our method excels in handling complex and diverse scenes without extensive training or additional 3D and multiview data. It leverages widely available pretrained NVS models for weak guidance, integrating this knowledge into a 3D-free view synthesis approach to achieve the desired results. Experimental results demonstrate that our method outperforms existing models in both qualitative and quantitative evaluations, providing high-fidelity and consistent novel view synthesis at desired camera angles across a wide variety of scenes., Comment: 13 pages, 12 figures, v3: corrected typos in figures

A significant part of contemporary research in autonomous vehicles is dedicated to the development of safety critical systems where state-of-the-art artificial intelligence (AI) algorithms, like computer vision (CV), can play a major role. Vision models have great potential for the real-time detection of numerous traffic signs and obstacles, which is essential to avoid accidents and protect human lives. Despite vast potential, computer vision-based systems have critical safety concerns too if the traffic condition drifts over time. This paper represents an analysis of how data drift can affect the performance of vision models in terms of traffic sign detection. The novelty in this research is provided through a YOLO-based fusion model that is trained with drifted data from the CARLA simulator and delivers a robust and enhanced performance in object detection. The enhanced model showed an average precision of 97.5\% compared to the 58.27\% precision of the original model. A detailed performance review of the original and fusion models is depicted in the paper, which promises to have a significant impact on safety-critical automotive systems.

Full waveform inversion (FWI) is a powerful tool for reconstructing material fields based on sparsely measured data obtained by wave propagation. For specific problems, discretizing the material field with a neural network (NN) improves the robustness and reconstruction quality of the corresponding optimization problem. We call this method NN-based FWI. Starting from an initial guess, the weights of the NN are iteratively updated to fit the simulated wave signals to the sparsely measured data set. For gradient-based optimization, a suitable choice of the initial guess, i.e., a suitable NN weight initialization, is crucial for fast and robust convergence. In this paper, we introduce a novel transfer learning approach to further improve NN-based FWI. This approach leverages supervised pretraining to provide a better NN weight initialization, leading to faster convergence of the subsequent optimization problem. Moreover, the inversions yield physically more meaningful local minima. The network is pretrained to predict the unknown material field using the gradient information from the first iteration of conventional FWI. In our computational experiments on two-dimensional domains, the training data set consists of reference simulations with arbitrarily positioned elliptical voids of different shapes and orientations. We compare the performance of the proposed transfer learning NN-based FWI with three other methods: conventional FWI, NN-based FWI without pretraining and conventional FWI with an initial guess predicted from the pretrained NN. Our results show that transfer learning NN-based FWI outperforms the other methods in terms of convergence speed and reconstruction quality.

Let $\mathcal C$ be a Grothendieck category and $U$ be a monad on $\mathcal C$ that is exact and preserves colimits. In this article, we prove that every hereditary torsion theory on the Eilenberg-Moore category of modules over a monad $U$ is differential. Further, if $\delta:U\longrightarrow U$ denotes a derivation on a monad $U$, then we show that every $\delta$-derivation on a $U$-module $M$ extends uniquely to a $\delta$-derivation on the module of quotients of $M$.

Magnetotransport measurements are crucial for understanding the Fermi surface properties, magnetism, and topology in quantum materials. Here, we report the discovery of giant room temperature odd-parity magnetoresistance (OMR) in a bilayer graphene (BLG) heterostructure interfaced with Cr$_2$Te$_2$Ge$_6$ (CGT). Using magnetotransport measurements, we demonstrate that the BLG/CGT heterostructure exhibits a significant antisymmetric longitudinal magnetoresistance, indicative of intrinsic time-reversal symmetry (TRS) breaking in the system. We show that the OMR is tunable via electrostatic gating. Additionally, the OMR is pronounced near the band edges and diminishes with increasing charge carrier density in graphene. Our theoretical analysis reveals that this phenomenon arises from the coupling of the out-of-plane components of Berry curvature and orbital magnetic moment to the applied magnetic field in a TRS-broken system. Our findings establish OMR as a significant probe for TRS breaking in quantum materials in which the crystal symmetries preclude the appearance of anomalous Hall effect., Comment: 34 pages

Distributed execution of deep learning training involves a dynamic interplay between hardware accelerator architecture and device placement strategy. This is the first work to explore the co-optimization of determining the optimal architecture and device placement strategy through novel algorithms, improving the balance of computational resources, memory usage, and data distribution. Our architecture search leverages tensor and vector units, determining their quantity and dimensionality, and on-chip and off-chip memory configurations. It also determines the microbatch size and decides whether to recompute or stash activations, balancing the memory footprint of training and storage size. For each explored architecture configuration, we use an Integer Linear Program (ILP) to find the optimal schedule for executing operators on the accelerator. The ILP results then integrate with a dynamic programming solution to identify the most effective device placement strategy, combining data, pipeline, and tensor model parallelism across multiple accelerators. Our approach achieves higher throughput on large language models compared to the state-of-the-art TPUv4 and the Spotlight accelerator search framework. The entire source code of PHAZE is available at https://github.com/msr-fiddle/phaze., Comment: Accepted at the 41st International Conference on Machine Learning (ICML), 2024

Deep learning kernels exhibit predictable memory accesses and compute patterns, making GPUs' parallel architecture well-suited for their execution. Software and runtime systems for GPUs are optimized to better utilize the stream multiprocessors, on-chip cache, and off-chip high-bandwidth memory. As deep learning models and GPUs evolve, access to newer GPUs is often limited, raising questions about the performance of new model architectures on existing GPUs, existing models on new GPUs, and new model architectures on new GPUs. To address these questions, we introduce NeuSight, a framework to predict the performance of various deep learning models, for both training and inference, on unseen GPUs without requiring actual execution. The framework leverages both GPU hardware behavior and software library optimizations to estimate end-to-end performance. Previous work uses regression models that capture linear trends or multilayer perceptrons to predict the overall latency of deep learning kernels on GPUs. These approaches suffer from higher error percentages when forecasting performance on unseen models and new GPUs. Instead, NeuSight decomposes the prediction problem into smaller problems, bounding the prediction through fundamental performance laws. NeuSight decomposes a single deep learning kernel prediction into smaller working sets called tiles, which are executed independently on the GPU. Tile-granularity predictions are determined using a machine learning approach and aggregated to estimate end-to-end latency. NeuSight outperforms prior work across various deep learning workloads and the latest GPUs. It reduces the percentage error from 198% and 19.7% to 3.8% in predicting the latency of GPT3 model for training and inference on H100, compared to state-of-the-art prior works, where both GPT3 and H100 were not used to train the framework.

Electric field-induced giant resistive switching triggered by insulator-to-metal transition (IMT) is one of the promising approaches for developing a new class of electronics often referred to as Mottronics. Achieving this resistive switching by minimal external field at room temperature is of paramount research and technological interest. Mott-IMT is often associated with structural modification, which is very important for optoelectronic and actuator applications. Here, we report a giant resistive switching of about 900 % at room temperature in disordered polycrystalline V2O3-Si thin film stabilized at the IMT phase boundary and associated structural transformation under a small electric field. The increase of electron population in the a1g band under the field is responsible for the Mott gap collapse that drives the structural transition. Furthermore, we also fabricated a room temperature Mott-FET with a channel ON/OFF resistive ratio of about 15. This study provides a fundamental mechanism of the Mott-IMT in V2O3 as well as its device applications., Comment: 16 pages, 5 figures

Cancer is a leading cause of death worldwide, necessitating advancements in early detection and treatment technologies. In this paper, we present a novel and highly efficient melanoma detection framework that synergistically combines the strengths of U-Net for segmentation and EfficientNet for the classification of skin images. The primary objective of our study is to enhance the accuracy and efficiency of melanoma detection through an innovative hybrid approach. We utilized the HAM10000 dataset to meticulously train the U-Net model, enabling it to precisely segment cancerous regions. Concurrently, we employed the ISIC 2020 dataset to train the EfficientNet model, optimizing it for the binary classification of skin cancer. Our hybrid model demonstrates a significant improvement in performance, achieving a remarkable accuracy of 99.01% on the ISIC 2020 dataset. This exceptional result underscores the superiority of our approach compared to existing model structures. By integrating the precise segmentation capabilities of U-Net with the advanced classification prowess of EfficientNet, our framework offers a comprehensive solution for melanoma detection. The results of our extensive experiments highlight the high accuracy and reliability of our method in both segmentation and classification tasks. This indicates the potential of our hybrid approach to significantly enhance cancer detection, providing a robust tool for medical professionals in the early diagnosis and treatment of melanoma. We believe that our framework can set a new benchmark in the field of automated skin cancer detection, encouraging further research and development in this crucial area of medical imaging.

To operate in open-ended environments where humans interact in complex, diverse ways, autonomous robots must learn to predict their behaviour, especially when that behavior is potentially dangerous to other agents or to the robot. However, reducing the risk of accidents requires prior knowledge of where potential collisions may occur and how. Therefore, we propose to gain this information by analyzing locations and speeds that commonly correspond to high-risk interactions within the dataset, and use it within training to generate better predictions in high risk situations. Through these location-based and speed-based re-weighting techniques, we achieve improved overall performance, as measured by most-likely FDE and KDE, as well as improved performance on high-speed vehicles, and vehicles within high-risk locations. 2023 IEEE International Conference on Robotics and Automation (ICRA)

As deep neural networks (DNNs) are increasingly used in safety-critical applications, there is a growing concern for their reliability. Even highly trained, high-performant networks are not 100% accurate. However, it is very difficult to predict their behavior during deployment without ground truth. In this paper, we provide a comparative and replicability study on recent approaches that have been proposed to evaluate the reliability of DNNs in deployment. We find that it is hard to run and reproduce the results for these approaches on their replication packages and even more difficult to run them on artifacts other than their own. Further, it is difficult to compare the effectiveness of the approaches, due to the lack of clearly defined evaluation metrics. Our results indicate that more effort is needed in our research community to obtain sound techniques for evaluating the reliability of neural networks in safety-critical domains. To this end, we contribute an evaluation framework that incorporates the considered approaches and enables evaluation on common benchmarks, using common metrics.

As more clinical workflows continue to be augmented by artificial intelligence (AI), AI literacy among physicians will become a critical requirement for ensuring safe and ethical AI-enabled patient care. Despite the evolving importance of AI in healthcare, the extent to which it has been adopted into traditional and often-overloaded medical curricula is currently unknown. In a scoping review of 1,699 articles published between January 2016 and June 2024, we identified 18 studies which propose guiding frameworks, and 11 studies documenting real-world instruction, centered around the integration of AI into medical education. We found that comprehensive guidelines will require greater clinical relevance and personalization to suit medical student interests and career trajectories. Current efforts highlight discrepancies in the teaching guidelines, emphasizing AI evaluation and ethics over technical topics such as data science and coding. Additionally, we identified several challenges associated with integrating AI training into the medical education program, including a lack of guidelines to define medical students AI literacy, a perceived lack of proven clinical value, and a scarcity of qualified instructors. With this knowledge, we propose an AI literacy framework to define competencies for medical students. To prioritize relevant and personalized AI education, we categorize literacy into four dimensions: Foundational, Practical, Experimental, and Ethical, with tailored learning objectives to the pre-clinical, clinical, and clinical research stages of medical education. This review provides a road map for developing practical and relevant education strategies for building an AI-competent healthcare workforce., Comment: 25 pages, 2 figures, 3 tables

Estimating the landscape and soil freeze-thaw (FT) dynamics in the Northern Hemisphere is crucial for understanding permafrost response to global warming and changes in regional and global carbon budgets. A new framework is presented for surface FT-cycle retrievals using L-band microwave radiometry based on a deep convolutional autoencoder neural network. This framework defines the landscape FT-cycle retrieval as a time series anomaly detection problem considering the frozen states as normal and thawed states as anomalies. The autoencoder retrieves the FT-cycle probabilistically through supervised reconstruction of the brightness temperature (TB) time series using a contrastive loss function that minimizes (maximizes) the reconstruction error for the peak winter (summer). Using the data provided by the Soil Moisture Active Passive (SMAP) satellite, it is demonstrated that the framework learns to isolate the landscape FT states over different land surface types with varying complexities related to the radiometric characteristics of snow cover, lake-ice phenology, and vegetation canopy. The consistency of the retrievals is evaluated over Alaska, against in situ ground-based observations, showing reduced uncertainties compared to the traditional methods that use thresholding of the normalized polarization ratio.