S
Shalabh Bhatnagar
Researcher at Indian Institute of Science
Publications - 308
Citations - 5153
Shalabh Bhatnagar is an academic researcher from Indian Institute of Science. The author has contributed to research in topics: Stochastic approximation & Markov decision process. The author has an hindex of 30, co-authored 294 publications receiving 4300 citations. Previous affiliations of Shalabh Bhatnagar include University of Marne-la-Vallée & Indian Institutes of Technology.
Papers
More filters
Proceedings Article
A generalized reduced linear program for markov decision processes
TL;DR: In this article, a generalized reduced linear program (GRLP) is defined, which has a tractable number of constraints that are obtained as positive linear combinations of the original constraints of the ALP.
Journal ArticleDOI
Analysis of Stochastic Approximation Schemes With Set-Valued Maps in the Absence of a Stability Guarantee and Their Stabilization
TL;DR: In this paper, it was shown that after a large number of iterations, if the stochastic approximation process enters the domain of attraction of an attracting set, it gets locked into the attracting set with high probability.
Book ChapterDOI
Kiefer-Wolfowitz Algorithm
TL;DR: This chapter reviews the Finite Difference Stochastic Approximation (FDSA) algorithm, and some of its variants for finding a local minimum of an objective function, and presents the original K-W scheme, first for the case of a scalar parameter, and subsequently for a vector parameter of arbitrary dimension.
Proceedings ArticleDOI
Model-based Safe Deep Reinforcement Learning via a Constrained Proximal Policy Optimization Algorithm
TL;DR: An On-policy Model-based Safe Deep RL algorithm in which an ensemble of neural networks with different initializations is used to tackle epistemic and aleatoric uncertainty issues faced during environment model learning and shows better reward performance than other constrained model-based approaches in the literature.
Book ChapterDOI
A Stochastic Approximation Algorithm for Quantile Estimation
TL;DR: Two new stochastic approximation algorithms for the problem of quantile estimation use the characterization of the quantile provided in terms of an optimization problem in [1] to take the shape of a stochastically gradient descent which minimizes the optimization problem.