My research interests broadly lie in the areas of reinforcement learning, online learning and stochastic optimization. Currently, my work focuses on designing low-complexity, provably efficient reinforcement learning algorithms for complex systems with large or continuous state and action spaces.

• (Under review) High-Probability Bounds for SGD under the Polyak-Lojasiewicz Condition with Markovian Noise.
  Avik Kar, Siddharth Chandak, Rahul Singh, Eric Moulines, Shalabh Bhatnagar, Nicholas Bambos.
  arXiv preprint arXiv:2603.14514.  [Link]
• Policy Zooming: Adaptive Discretization-based Infinite-Horizon Average-Reward Reinforcement Learning.
  Avik Kar and Rahul Singh.
  Accepted in the 40th AAAI Conference on Artificial Intelligence.  [Link]
• Provably Adaptive Average Reward Reinforcement Learning for Metric Spaces.
  Avik Kar and Rahul Singh.
  Accepted in the 41st Conference on Uncertainty in Artificial Intelligence (UAI).  [Link]
• Fantom: Federated Adversarial Network for Training Multi-Sequence MRI in Semantic Segmentation.
  Anupam Borthakur, Apoorva Srivastava, Avik Kar, Dipayan Dewan, and Debdoot Sheet.
  International Conference on Image Processing (ICIP). IEEE, 2024.  [Link]
• Linear Bandits With Side Observations on Networks.
  Avik Kar, Rahul Singh, Fang Liu, Xin Liu, and Ness B. Shroff.
  IEEE/ACM Transactions on Networking, 2024.  [Link]
• Finite Time Logarithmic Regret Bounds for Self-Tuning Regulation.
  Rahul Singh, Akshay Mete, Avik Kar, and P R Kumar.
  41st International Conference on Machine Learning (ICML), 2024.  [Link]
• Federated Learning for Site Aware Chest Radiograph Screening.
  Arunava Chakravarty, Avik Kar, Ramanathan Sethuraman, and Debdoot Sheet.
  18th International Symposium on Biomedical Imaging (ISBI). IEEE, 2021.  [Link]