I work in the area of Reinforcement Learning (RL), which focuses on finding an optimal policy in stochastic environments. My research centers on developing efficient algorithms for solving Bellman’s optimality equation and, more broadly, on improving methods across both model-based and model-free reinforcement learning. A key aspect of my work is the design of algorithms with rigorous theoretical convergence guarantees, ensuring both computational efficiency and robustness.
In addition to single-agent settings, I have recently begun exploring multi-agent reinforcement learning, with a particular interest in two-player zero-sum stochastic games and their connections to dynamic programming and game-theoretic solution concepts. Looking ahead, I am keen on applying these algorithmic frameworks to real-world problems, bridging theory with practical impact.
Shreyas, S. R. Double Successive Over-Relaxation Q-Learning with an Extension to Deep Reinforcement Learning. IEEE Transactions on Neural Networks and Learning Systems, 36, 19467–19472 (2025). DOI: https://doi.org/10.1109/TNNLS.2025.3576581
Shreyas, S. R. and Antony Vijesh. A Multi-Step Minimax Q-Learning Algorithm for Two-Player Zero-Sum Markov Games. Neurocomputing, 657, Article 131552 (2025). DOI: https://doi.org/10.1016/j.neucom.2025.131552
Antony Vijesh, S. R. Shreyas, and M. S. Abdulla. A Note on Generalized Second-Order Value Iteration in Markov Decision Processes. Journal of Optimization Theory and Applications, 199, 1022–1049 (2023). DOI: https://doi.org/10.1007/s10957-023-02309-x
Antony Vijesh and S. R. Shreyas. A Weighted Smooth Q-Learning Algorithm. IEEE Control Systems Letters, 9, 21–26 (2025). DOI: https://doi.org/10.1109/LCSYS.2025.3551265
A. Sharma, S. R. Shreyas, and Antony Vijesh. Order of Convergence Estimate for a Stochastic Iterative Method for the Mode and Its Size. Communications in Statistics—Simulation and Computation ,1–15, (2026). DOI: https://doi.org/10.1080/03610918.2026.2671369.
S. R. Shreyas and M. Joseph. Characterization of Signed Paths and Cycles Admitting Minus Dominating Function. Communications in Combinatorics and Optimization, 5, 61–68 (2020). Available at: https://comb-opt.azaruniv.ac.ir/article_13977.html
Completed an internship at Indian Institute of Science, Bengaluru, from April 2017 to June 2017
Participated in the IEEE Computational Intelligence Summer School on Deep Learning and Computational Intelligence: Theory and Applications at Indian Institute of Technology Indore in May 2022.
Attended the workshop AI-MPOWER: Empowering Educators for the Data-Driven Future at Indian Institute of Science in November 2023.
Presented my research work at the ICC Conference 2025, organized by Indian Institute of Science.