Skip to the content
- April 2026: Our work titled "Corruption-Tolerant Asynchronous Q-Learning with Near-Optimal Rates" accepted at the 43rd International Conference on Machine Learning (ICML 2026).
- March 2026: Invited talk "Towards a Finite-Time Theory for Adversarially-Robust Reinforcement Learning" delivered at the CORAL Seminar, ECE Dept., NCSU.
- January 2026: Our works titled "Robust Federated Q-Learning with Almost No Communication", and "Variance-Reduced Q-Learning for Static and Time-Varying Networks" accepted for "Oral" presentation, and at publication at 2026 American Control Conference.
- October 2025: Presented a Poster titled "Robust Federated Reinforcement Learning with Byzantine Agents" at the Applied AI Symposium, 2025.
- September 2025: Our recent work titled "Corruption-Tolerant Asynchronous Q-Learning with Near-Optimal Rates" accepted at 39th Annual Conference on Neural Information Processing Systems (NeurIPS 2025)-Reliable ML Workshop.
- September 2025: Our recent work titled "Corruption-Tolerant Asynchronous Q-Learning with Near-Optimal Rates" is released on Arxiv. The work provides first provably robust variants of asynchronous Q-learning that tolerates adversarially corrupted rewards. The algorithm is distribution-agnostic, and achieves near-optimal finite-time guarantees up to a provably unavoidable corruptive term.
- September 2025: Poster "Adversarially-Robust TD Learning with Markovian Data" accepted at the New York Reinforcement Learning Workshop 2025, Amazon.
- April 2025: Passed my Ph.D. Oral Qualifying Examination! The committee members were my advisor Dr. Aritra Mitra (Assistant Professor, Dept. of ECE) , Dr. Wenbin Lu (Professor, Director of Graduate Programs, Dept. of Statistics), Dr. Aranya Chakrabortty (IEEE Fellow, Professor, Associate Dept. Head for Research, Dept. of ECE), and Dr. Edgar Lobaton (Professor, Dept. of ECE).
- April 2025: Poster "Towards Finite-Time Rates for Adversarially -Robust Reinforcement Learning" accepted at the Northeast Systems and Control Workshop 2025, Columbia University.
- February 2025: Our extended pre-print titled "Adversarially-Robust TD Learning with Markovian Data: Finite-Time Rates and Fundamental Limits" is released on ArXiv. (To appear in 28th International Conference on Artificial Intelligence and Statistics (AISTATS) 2025).
- January 2025: Our paper titled"Adversarially-Robust TD Learning with Markovian Data: Finite-Time Rates and Fundamental Limits" accepted at 28th International Conference on Artificial Intelligence and Statistics (AISTATS) 2025.
- December 2024: Presented our paper "Robust Q-Learning under Corrupted Rewards" in the 63rd Conference on Decision and Control 2024 at Milan, Italy.
- September 2024: Presented a Poster titled "Robust Algorithms for Adversarial Reinforcement Learning" at the Applied AI Symposium, Theoretical Machine Learning Track.
- September 2024: Our extended pre-print titled "Robust Q-Learning under Corrupted Rewards" is released on ArXiv. (To appear in 63rd Conference on Decision and Control 2024).
- September 2024: Received the "IEEE CDC 24 Student Travel Support Award".
- July 2024: Our paper titled "Robust Q-Learning under Corrupted Rewards" accepted at the flagship 63rd IEEE Conference on Decision and Control 2024.
- August 2023 : Joined North Carolina State University, Raleigh as a Ph.D. Student under Dr. Aritra Mitra.
- July 2023: Graduated from Indian Institute of Science, Bangalore (M.Tech Robotics and Autonomous Systems).
- February - March 2023: Received Ph.D. offers from North Carolina State University, Purdue University.