I will be in Job Market 2023-2024!!
I am currently a postdoctoral researcher at the Halıcıoğlu Data Science Institute, University of California, San Diego. I work primarily with Prof. Yian Ma. I finished my Ph.D. in June 2020, and my advisor were Prof. Prasant Mohapatra and Prof. Krishna Balasubramanian. From 2020-2022, I was a Postdoc at the Department of Statistics, UC Davis working with Prof. Krishna Balasubramanian. Prior to this, I received a Bachelor of Technology (Hons.) in Electronics and Electrical Communication Engineering in 2013 from the Indian Institute of Technology, Kharagpur.
I am interested in non-convex optimization, uncertainty quantification, Markov Chain Monte Carlo (MCMC) sampling, multi-objective optimization, and robust learning from dependent data. Here is a link to my google scholar account.
NEWS
I will be attending INFORMS Annual Meet
New work Optimization on Pareto sets: On a theory of multi-objective optimization
New work Online covariance estimation for stochastic gradient descent under Markovian sampling
Our paper Fairness Uncertainty Quantification: How certain are you that the model is fair? featured on Montreal AI Ethics Institute newsletter
New paper A Central Limit Theorem for Stochastic Saddle Point Optimization
New paper Fairness Uncertainty Quantification: How certain are you that the model is fair?
Giving a talk on Constrained Stochastic Nonconvex Optimization with State-dependent Markov Data at SIAM Conference on Optimization (OP23)
"Constrained Stochastic Nonconvex Optimization with State-dependent Markov Data" ( with Krishnakumar Balasubramanian, and Saeed Ghadimi) got accepted at NeurIPS 2022.
Excited to work with Misha Belkin and Yian Ma at HDSI as a postdoc from August, 2022!
"Stochastic Zeroth-Order Optimization under Nonstationarity and Nonconvexity", ( with Krishnakumar Balasubramanian, Saeed Ghadimi, and Prasant Mohapatra), JMLR (to appear), 2022.
"On Empirical Risk Minimization with Dependent and Heavy-Tailed Data" ( with Krishnakumar Balasubramanian, Murat A. Erdogdu) got accepted at NeurIPS 2021.
I am attending Sampling Algorithms and Geometries on Probability Distributions
I am attending Geometric Methods in Optimization and Sampling Boot Camp
"Stochastic Zeroth-order Discretizations of Langevin Diffusions for Bayesian Inference", ( with Lingqing Shen, Krishnakumar Balasubramanian, and Saeed Ghadimi), Bernoulli (to appear), 2021.
PUBLICATIONS
Journal
Abhishek Roy, Lingqing Shen, Krishnakumar Balasubramanian, and Saeed Ghadimi. "Stochastic Zeroth-order Discretizations of Langevin Diffusions for Bayesian Inference", Bernoulli 2021 (to appear).
Abhishek Roy, Krishnakumar Balasubramanian, Saeed Ghadimi, and Prasant Mohapatra. "Stochastic Zeroth-Order Optimization under Nonstationarity and Nonconvexity", JMLR 2021 (to appear).
Conference
Abhishek Roy, Krishnakumar Balasubramanian, and Saeed Ghadimi. "Constrained Stochastic Nonconvex Optimization with State-dependent Markov Data", (Accepted at NeurIPS 2022).
Abhishek Roy, Krishnakumar Balasubramanian, and Murat A. Erdogdu. "On Empirical Risk Minimization with Dependent and Heavy-Tailed Data.", (Accepted at NeurIPS 2021).
Abhishek Roy, Krishnakumar Balasubramanian, Saeed Ghadimi, and Prasant Mohapatra, “Escaping Saddle-Point Faster under Interpolation-like Conditions”, (Accepted at NeurIPS 2020).
Abhishek Roy*, Anshuman Chhabra*, and Prasant Mohapatra. “Suspicion-Free Adversarial Attacks on Clustering Algorithms.” (Accepted at AAAI 2020).
Abhishek Roy, Charles A. Kamhoua†, and Prasant Mohapatra, “Game Theoretic Characterization of Collusive Behavior among Attackers”, IEEE International Conference on Computer Communications (INFOCOM), Honolulu, Hawaii, Apr. 2018.
Preprint
Abhishek Roy*, Geelon So*, and Yi-An Ma. "Optimization on Pareto sets: On a theory of multi-objective optimization", 2023 (under review).
Abhishek Roy, and Krishnakumar Balasubramanian. "Online covariance estimation for stochastic gradient descent under Markovian sampling", 2023.
Abhishek Roy, and Yi-An Ma. "A Central Limit Theorem for Stochastic Saddle Point Optimization", 2023 (under review).
Abhishek Roy, and Prasant Mohapatra. "Fairness Uncertainty Quantification: How certain are you that the model is fair?", 2023 (under review).
Abhishek Roy, Yifang Chen, Krishnakumar Balasubramanian, and Prasant Mohapatra. "Online and Bandit Algorithms for Nonstationary Stochastic Saddle-Point Optimization.", 2019.
Abhishek Roy, Krishnakumar Balasubramanian, Saeed Ghadimi, and Prasant Mohapatra. "Multi-Point Bandit Algorithms for Nonstationary Online Nonconvex Optimization.", 2019.
TALKS
TEACHING
HONORS & AWARDS
Infocom 2018 Best-in-Session Presentation Award
ECE Graduate Program Travel Award, University of California, Davis, 2018
ECE Graduate Program Fellowship, University of California, Davis, 2015
University of Tokyo IIT Undergraduate Students Scholarship, 2013
Jagadis Bose National Science Talent Search Scholarship, 2009
EDUCATION
July 2013 - June 2020. Ph.D. (Electrical and Computer Engineering), University of California, Davis.
October 2016 - June 2017 M.S (Statistics), University of California, Davis
July 2009 - April 2013 B.Tech (Hons.) (Electronics and Electrical Communication Engineering), Indian Institute of
Technology, Kharagpur
INTERESTS & HOBBIES
Hiking
Mountains + hike = bliss. Sierra Nevada is my go-to place to destress.
Photography
I consider waiting hours for a frame I want, to be a good usage of time :) The background is from a Hawaii trip.