arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Samarth Gupta

30 papers · 2015–2025

2015: 12016: 32017: 12018: 42019: 22020: 72021: 22022: 22023: 32024: 22025: 3
2015peak 7 in 20202025

Frequent co-authors

Papers

showing 1–25 of 30 RSS

Bandit-based Communication-Efficient Client Selection Strategies for Federated Learning

Yae Jee Cho, Samarth Gupta, Gauri Joshi +1 more

cs.LG · · cited 95× · arXiv ↗ · more like this ·

Due to communication constraints and intermittent client availability in federated learning, only a subset of clients can participate in each training round. While most prior works assume uniform and unbiased client selection, recent work on biased client selection has shown that selecting clients with higher local losses can improve error convergence speed. However, previously proposed biased selection strategies either require additional communication cost for evaluating the exact local loss or utilize stale local loss, which can even make the model diverge. In this paper, we present a bandit-based communication-efficient client selection strategy UCB-CS that achieves faster convergence with lower communication overhead. We also demonstrate how client selection can be used to improve fairness.

Object Detection in Foggy Conditions by Fusion of Saliency Map and YOLO

Sarthak Katyal, Sanjay Kumar, Ronak Sakhuja +1 more

Computer Vision and Pattern Recognition · · cited 26× · DOI ↗ · more like this ·

Under foggy conditions, visibility decreases and causes many problems. Less visibility due to foggy conditions while driving increases the risk of road accidents. It is important to detect and recognize the nearby objects under such conditions and predict the distance of collision. There is a need to devise an object detection mechanism during foggy conditions. The paper proposes a solution to this problem by proposing a VESY(Visibility Enhancement Saliency YOLO) sensor which uses an algorithm that fuses the saliency map of the foggy image frame with the output generated from object detection algorithm YOLO (You Only Look Once). The image is sensed using image sensors in a stereo camera which are activated using a fog sensor and a depth map is generated to calculate the distance of collision. Dehaze algorithm is applied to improve the quality of the image frame whose Saliency image is…

A scalable parallel implementation of evolutionary algorithms for multi-objective optimization on GPUs

Samarth Gupta, Gary Tan

Computational Theory and Mathematics · · cited 26× · DOI ↗ · more like this ·

Multi-Objective Evolutionary Algorithms(MOEAs) have been gaining increased popularity and usage in different fields of engineering. For real world large scale optimization problems with large variable/search space, using a large population of individuals in proportion to the size of search space is ubiquitous. Solving such problems with current state of the art algorithms like NSGA-II [1] is pervasive. The strength of NSGA-II lies in its non-dominance selection procedure and non-dominance based sorting of a population of individuals. Although, the non-dominated sort is computationally efficient for a small population (10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> - 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> ) of solutions but becomes computationally expensive and…

Analysis on Symptoms Driven Disease Risk Assessment using Artificial Intelligence Approach

Anisha Sharma, Samarth Gupta, Sanjay Kumar Dubey

Artificial Intelligence · · cited 5× · DOI ↗ · more like this ·

An important part of healthcare is disease risk assessment, which aids in early diagnosis, prevention, and personalized treatment plans. Symptoms, as major indications of underlying health issues, provide critical information for appropriate risk assessment. This review paper investigates the junction of Symptom Driven Disease Risk Assessment using Artificial Intelligence, providing an in-depth examination of current advances and problems in this field. Through this holistic approach to disease prediction and personalized health recommendations, the AI-driven system aims to revolutionize healthcare by empowering individuals to proactively manage their health, leading to a healthier society and ultimately reducing the strain on healthcare systems. The intricacies of symptom analysis and disease prediction were discovered by comparing different approaches of machine learning as well deep…

Real Time Optimization of Network Control Strategies in DynaMIT2.0

Samarth Gupta, Ravi Seshadri, Bilge Atasoy +8 more

Transportation · Transportation Research Board 95th Annual MeetingTransportation Research Board · · cited 5× · OpenAlex ↗ · more like this ·

Traffic congestion is a pervasive problem in most urban transportation networks worldwide owing to significant supply and demand side uncertainty. Unfortunately, network capacity augmentation through infrastructure expansion is seldom a feasible solution due to financial, environmental and space restrictions. Consequently, emphasis is shifting towards alleviating network congestion through a more efficient utilization of existing infrastructure. In this context, this paper proposes a framework that integrates the real time optimization of network control strategies (tolls, ramp metering rates, etc.) with the generation of predictive travel time information within DynaMIT2.0, a state of the art Dynamic Traffic Assignment (DTA) model. The proposed approach is novel in that it takes into account the network state estimated using heterogeneous real time data from a variety of sources and…

User Specific Safe Route Recommendation System

Yash S Asawa, Samarth Gupta, Varun +1 more

Signal Processing · Open MIND · · cited 4× · DOI ↗ · more like this ·

Criminal activities have reached unprecedented levels in almost every part of the world. Desperate times like these require desperate measures to ensure the safety of people, especially people that need to travel on a daily basis to places, known and unknown. The majority of these criminal offences occur while the victim is travelling, irrespective of the mode of transport: walking, personal vehicles, public transport vehicles, auto-rickshaws, or cabs. This paper proposes a User-Specific Safe Route Recommendation System which presents a safe route visualized on maps to the user based on the past criminal records of the geographical region. Our approach is implemented on two-levels, first to realize the user-specific features using a Decision Network and the latter to actuate the safe route generation using Geospatial Data Analysis. We visualize the determined routes through a…

User Targeted Offline Advertising using Recognition Based Demographics and Queue Scheduling

Ruchika Malhotra, Samarth Gupta, Sarthak Katyal +1 more

Computer Vision and Pattern Recognition · International Journal of Engineering and Advanced Technology · · cited 4× · DOI ↗ · more like this ·

Offline advertisements are static in nature. Advertising companies use billboards for advertising. These billboards display advertisements in a random fashion depending on the investment made by the advertiser. Advertisers pay a fixed amount of money for displaying their advertisements and not on the basis of relevant viewership. The technology proposed in the paper ensures that this disparity is handled wherein offline advertisements are targeted to the relevant audience. The technology has been named TARP which is an abbreviation for Target. Advertise. Revolutionise. Promote. TARP uses built in cameras on offline advertising platforms such as billboards &amp; TV Screens in malls, restaurants, metro &amp; airports to target advertisements based on gender, age and other relevant demographics. The technology is a boon for the advertising industry and benefits both advertisers and…

Approximate Newton policy gradient algorithms

Haoya Li, Samarth Gupta, Hsiangfu Yu +2 more

cs.LG · · cited 3× · arXiv ↗ · more like this ·

Policy gradient algorithms have been widely applied to Markov decision processes and reinforcement learning problems in recent years. Regularization with various entropy functions is often used to encourage exploration and improve stability. This paper proposes an approximate Newton method for the policy gradient algorithm with entropy regularization. In the case of Shannon entropy, the resulting algorithm reproduces the natural policy gradient algorithm. For other entropy functions, this method results in brand-new policy gradient algorithms. We prove that all these algorithms enjoy Newton-type quadratic convergence and that the corresponding gradient flow converges globally to the optimal solution. We use synthetic and industrial-scale examples to demonstrate that the proposed approximate Newton method typically converges in single-digit iterations, often orders of magnitude faster…

Real Time Face Recognition on an Edge Computing Device

Samarth Gupta

Computer Vision and Pattern Recognition · · cited 3× · DOI ↗ · more like this ·

Face recognition systems have vast applications in surveillance systems and human-computer interactions. Different approaches such as Principal Component Analysis, Fisher linear discriminant analysis, Convolutional Neural Networks (CNN) have been commonly used for face recognition. However, in the recent times, CNN's have shown quite promising results in various face recognition systems. But, deep learning based CNNs have many limitations such as they require extensive training data, have excessively high computational and cooling requirements, and lack flexibility in deployment. Fields such as robotics and embedded systems that deploy face recognition systems have significantly less power on board and limited heat dissipation capacity. Therefore, it becomes difficult to deploy deep learning models on them but edge computing based devices like the Intel Neural Stick bridge this gap as…

Effect of Recommendations on Serving Content with Unknown Demand

Samarth Gupta, Sharayu Moharir

Computer Networks and Communications · · cited 3× · DOI ↗ · more like this ·

We consider the task of content replication in distributed content delivery systems used by Video-on-Demand (VoD) services with large content catalogs. The prior work in this area focuses on the setting where each request is generated independent of all past requests. Motivated by the fact that most popular VoD services offer recommendations to users based on their viewing history, in a departure from existing studies, we study the setting where there is time-correlation in requests coming from each user.

Screening and Identification of Natural Compounds as Potential Inhibitors of Glutamate Racemase, an Emerging Drug Target of Food Pathogen E. coli O157:H7: An In-silico Approach to Combat Increasing Drug Resistance

Rajnish Kumar, Samarth Gupta, Sujata Adhana +7 more

Computational Theory and Mathematics · Infectious Disorders - Drug Targets · · cited 2× · DOI ↗ · more like this ·

BACKGROUND: ) O157:H7, capable of causing serious food-borne illnesses, is extensively studied and is known to be transmitted through animal reservoirs or person-to-person contact, leading to severe disease outbreaks. The emergence of antibiotic resistance in these strains, coupled with increased adverse effects of existing therapeutics, underscores the urgent need for alternative therapeutic strategies. OBJECTIVE: O157:H7 (EC MurI) as a novel drug target. Furthermore, the study seeks to identify new compounds with potential inhibitory effects against this protein. METHODS: Using computational tools, the study identified inhibitor binding sites on EC MurI and identified relevant inhibitors capable of binding to these sites. Molecular docking techniques were employed to assess potential hits, and selected compounds were further analyzed for their structural activity and binding affinity…

Uncertainty Informed Optimal Resource Allocation with Gaussian Process based Bayesian Inference

Samarth Gupta, Saurabh Amin

math.OC · · cited 1× · arXiv ↗ · more like this ·

We focus on the problem of uncertainty informed allocation of medical resources (vaccines) to heterogeneous populations for managing epidemic spread. We tackle two related questions: (1) For a compartmental ordinary differential equation (ODE) model of epidemic spread, how can we estimate and integrate parameter uncertainty into resource allocation decisions? (2) How can we computationally handle both nonlinear ODE constraints and parameter uncertainties for a generic stochastic optimization problem for resource allocation? To the best of our knowledge current literature does not fully resolve these questions. Here, we develop a data-driven approach to represent parameter uncertainty accurately and tractably in a novel stochastic optimization problem formulation. We first generate a tractable scenario set by estimating the distribution on ODE model parameters using Bayesian inference…

Massive Open Online Courses

Shikha Gupta, Samarth Gupta

Computer Science Applications · Advances in educational technologies and instructional design book series · · cited 1× · DOI ↗ · more like this ·

Multimedia technology and the internet have revolutionized the delivery and the reach of education through massive open online courses (MOOCs). Starting in 2011 when professors from Stanford University took a lead in starting such courses, teaching/learning through MOOCs has become a revolution of sorts with the professors and the higher education institutions (HEIs) realizing the benefits of several thousand students registering for an online course. Today, more than 11000 MOOCs are available from various countries spanning diverse cultures and languages, disrupting the teaching/learning models in the HEIs. This chapter outlines the history of MOOCs. It also suggests research questions towards the use of MOOCs in promoting international/intercultural communication. A critical assessment of the impact of online learning and MOOCs in the COVID-19 era is also presented.

Request Patterns and Caching for VoD Services with Recommendation Systems

Samarth Gupta, Sharayu Moharir

cs.NI · · cited 1× · arXiv ↗ · more like this ·

Video on Demand (VoD) services like Netflix and YouTube account for ever increasing fractions of Internet traffic. It is estimated that this fraction will cross 80% in the next three years. Most popular VoD services have recommendation engines which recommend videos to users based on their viewing history, thus introducing time-correlation in user requests. Understanding and modeling this time-correlation in user requests is critical for network traffic engineering. The primary goal of this work is to use empirically observed properties of user requests to model the effect of recommendation engines on the request patterns in VoD services. We propose a Markovian request model to capture the time-correlation in user requests and show that our model is consistent with the observations of existing empirical studies. Most large-scale VoD services deliver content to users via a distributed…

Bayesian Optimization in High Dimensional Input Space

Jaemin Son, Samarth Gupta, Gary Tan

Computational Theory and Mathematics · Simulation Tools and Techniques for Communications, Networks and System · · cited 1× · DOI ↗ · more like this ·

Many simulations include design parameters that influence the terminal outcome. Often, our aim is to optimize these design parameters through recursive simulation in order to gain the outcome in favor of our interests. However, unlike usual objective functions, simulation cannot be directly expressed in analytic forms. Thus, conventional optimization methods such as gradient descent and convex optimization do not apply straightforwardly. Also, simulation usually incurs high computational costs. Therefore, less number of function evaluations is preferred during the optimization process. We present Bayesian Optimization (BO) to deal with such challenges in simulation-based optimization. BO constructs surrogates of the true function and evaluates the most promising points based on the surrogates. Also, we consider issues in applying BO to problems with high dimensional input space. Dealing…

Disentanglement in T-space for Faster and Distributed Training of Diffusion Models with Fewer Latent-states

Samarth Gupta, Raghudeep Gadde, Rui Chen +1 more

cs.LG · · cited 0× · arXiv ↗ · more like this ·

We challenge a fundamental assumption of diffusion models, namely, that a large number of latent-states or time-steps is required for training so that the reverse generative process is close to a Gaussian. We first show that with careful selection of a noise schedule, diffusion models trained over a small number of latent states (i.e. $T \sim 32$) match the performance of models trained over a much large number of latent states ($T \sim 1,000$). Second, we push this limit (on the minimum number of latent states required) to a single latent-state, which we refer to as complete disentanglement in T-space. We show that high quality samples can be easily generated by the disentangled model obtained by combining several independently trained single latent-state models. We provide extensive experiments to show that the proposed disentangled model provides 4-6$\times$ faster convergence…

Codex comment: lexical and syntactic enhanced code comment generation

Pankaj Ganage, S.K. Mahajan, Sachin B. Jadhav +3 more

Artificial Intelligence · International Journal of Advanced Mathematical Sciences · · cited 0× · DOI ↗ · more like this ·

Insufficient or poor-quality code comments are the result of inexperience or intentional omission, which can ‎hinder software comprehension and pose challenges, particularly in the context of maintenance within a ‎team-programming environment. Here, the smart tools for a more detailed understanding, such as ‎automatic generation of comprehensible code comments, can give insights about functionality and purpose of ‎every element of the code. During the last few years, studies were conducted about this problem since code ‎comments are the only way for human constructively to code is to be conveyed and are therefore highly ‎important to developers. The most popular solutions are inspired by recent breakthroughs in NMT in casting ‎the problem of generating code comments as an NMT problem consisting of translating code into sounding-‎the-best-of-human comments. With such a level of…

APPORTION- THE BILL SPLIT

Moksha Jain, Naman Tenguria, Rochak Sharma +4 more

Computer Networks and Communications · International Research Journal of Modernization in Engineering Technology and Science · · cited 0× · DOI ↗ · more like this ·

The project describes and presents a system designed and built with the intention of reducing paper-based transactions by taking advantage of the technology and recent legislature which allows the service to eliminate deficiencies associated with financial transactions settled via paper-based means.Apportion-Bill Split is a payment flexible system that can be very useful at the time of group events when several members are included and each of them has performed some sort of transaction for the group.At the end of the event, it might be a bit chaotic to calculate who owns how much money to other members.To sort this issue, Apportion-Bill Spit is coined which will allow the user to add the expenses made by him and with whom the expense should be equally shared.

Spotlight in Plastic Surgery: April 2023

Brett T. Phillips, Ronald K. Akiki, Amanda Fazzalari +9 more

Gender Studies · Plastic & Reconstructive Surgery · · cited 0× · DOI ↗ · more like this ·

“Spotlight in Plastic Surgery” provides a quarterly overview of articles from non–plastic surgery high-impact journals and Plastic and Reconstructive Surgery Global Open. This month, we review key articles of relevance to plastic surgeons from The Lancet, New England Journal of Medicine, Nature, Annals of Surgery, JAMA Surgery, and Plastic and Reconstructive Surgery Global Open. Our goal is to enlighten Journal readers with clinical and basic science literature relevant to our specialty that they otherwise may not be read. We continue to thank the Plastic and Reconstructive Surgery resident advisors and advisory board members who volunteer to help us identify these articles and provide brief synopses. Should any readers have specific suggestions of articles to review in recent issues of high-impact journals, please contact us by email at [email protected] We wish to remain responsive to…

Bayesian regularization of empirical MDPs

Samarth Gupta, Daniel N. Hill, Lexing Ying +1 more

cs.LG · · cited 0× · arXiv ↗ · more like this ·

In most applications of model-based Markov decision processes, the parameters for the unknown underlying model are often estimated from the empirical data. Due to noise, the policy learnedfrom the estimated model is often far from the optimal policy of the underlying model. When applied to the environment of the underlying model, the learned policy results in suboptimal performance, thus calling for solutions with better generalization performance. In this work we take a Bayesian perspective and regularize the objective function of the Markov decision process with prior information in order to obtain more robust policies. Two approaches are proposed, one based on $L^1$ regularization and the other on relative entropic regularization. We evaluate our proposed algorithms on synthetic simulations and on real-world search logs of a large scale online shopping store. Our results demonstrate…

Integer Programming-based Error-Correcting Output Code Design for Robust Classification

Samarth Gupta, Saurabh Amin

cs.LG · · cited 0× · arXiv ↗ · more like this ·

Error-Correcting Output Codes (ECOCs) offer a principled approach for combining simple binary classifiers into multiclass classifiers. In this paper, we investigate the problem of designing optimal ECOCs to achieve both nominal and adversarial accuracy using Support Vector Machines (SVMs) and binary deep learning models. In contrast to previous literature, we present an Integer Programming (IP) formulation to design minimal codebooks with desirable error correcting properties. Our work leverages the advances in IP solvers to generate codebooks with optimality guarantees. To achieve tractability, we exploit the underlying graph-theoretic structure of the constraint set in our IP formulation. This enables us to use edge clique covers to substantially reduce the constraint set. Our codebooks achieve a high nominal accuracy relative to standard codebooks (e.g., one-vs-all, one-vs-one, and…

Effect of Entrepreneurial Mindset on Funding Opportunities for Fundamental Research

Samarth Gupta, Eunhye Kim, Greg Strimel

Communication · Papers on Engineering Education Repository (American Society for Engineering Education) · · cited 0× · DOI ↗ · more like this ·

Abstract Proliferation of Lean Launchpad, a curriculum designed by Steve Blank and adopted by National Science Foundation as Innovation-Corps program, has created myriad opportunities for commercialization of existing technology and funding innovative ideas which otherwise would never have seen the light of the day. Identifying the merits of funding research based on value proposition and customer demand, key components of the Business Model Canvas, has created a paradox for Endowment Funds, Family Offices and even Federal Funding agencies on whether to continue support for fundamental research in hi-tech laboratories or to think like a Venture Capital investor hoping to hit a home-run. In a world where the focus is on market viability of research, is the future bleak for fundamental scientific curiosity? Are we on the right path turning Endowment Funds, Family Offices and Federal…

Supervised machine learning for region assignment of zebrafish brain nuclei based on computational assessment of cell neighborhoods

Samarth Gupta, Yuan Xue, Yifu Ding +5 more

Computer Vision and Pattern Recognition · · cited 0× · DOI ↗ · more like this ·

Histological studies provide cellular insights into tissue architecture and have been central to phenotyping and biological discovery. Synchrotron X-ray micro-tomography of tissue, or “X-ray histotomography”, yields three-dimensional reconstruction of fixed and stained specimens without sectioning. These reconstructions permit the computational creation of histology-like sections in any user-defined plane and slice thickness. Furthermore, they provide an exciting new basis for volumetric, computational histological phenotyping at cellular resolution. In this paper, we demonstrate the computational characterization of the zebrafish central nervous system imaged by Synchrotron X-ray micro-CT through the classification of small cellular neighborhood volumes centered at each detected nucleus in a 3D tomographic reconstruction. First, we propose a deep learning-based nucleus detector to…

Real-Time Predictive Control Strategy Optimization

Samarth Gupta, Ravi Seshadri, Bilge Atasoy +4 more

eess.SY · · cited 0× · arXiv ↗ · more like this ·

Traffic congestion has lead to an increasing emphasis on management measures for a more efficient utilization of existing infrastructure. In this context, this paper proposes a novel framework that integrates real-time optimization of control strategies (tolls, ramp metering rates, etc.) with guidance generation using predicted network states for Dynamic Traffic Assignment systems. The efficacy of the framework is demonstrated through a fixed demand dynamic toll optimization problem which is formulated as a non-linear program to minimize predicted network travel times. A scalable efficient genetic algorithm is applied to solve this problem that exploits parallel computing. Experiments using a closed-loop approach are conducted on a large scale road network in Singapore to investigate the performance of the proposed methodology. The results indicate significant improvements in network…

Active Distribution Learning from Indirect Samples

Samarth Gupta, Gauri Joshi, Osman Yağan

cs.LG · · cited 0× · arXiv ↗ · more like this ·

This paper studies the problem of {\em learning} the probability distribution $P_X$ of a discrete random variable $X$ using indirect and sequential samples. At each time step, we choose one of the possible $K$ functions, $g_1, \ldots, g_K$ and observe the corresponding sample $g_i(X)$. The goal is to estimate the probability distribution of $X$ by using a minimum number of such sequential samples. This problem has several real-world applications including inference under non-precise information and privacy-preserving statistical estimation. We establish necessary and sufficient conditions on the functions $g_1, \ldots, g_K$ under which asymptotically consistent estimation is possible. We also derive lower bounds on the estimation error as a function of total samples and show that it is order-wise achievable. Leveraging these results, we propose an iterative algorithm that i) chooses the…