publications

Research on online learning, bandits, optimization, and machine learning theory.

2026

  1. Optimal Anytime Algorithms for Online Convex Optimization with Adversarial Constraints
    Dhruv Sarkar, and Abhishek Sinha
    In The 43rd International Conference on Machine Learning. Preliminary version appeared at the AAAI 2026 MATH4AI Workshop. , 2026
  2. Improved Algorithms for Nash Welfare in Linear Bandits
    Dhruv Sarkar, Nishant Pandey, and Sayak Ray Chowdhury
    In The 43rd International Conference on Machine Learning, 2026
  3. Projection-free Algorithms for Online Convex Optimization with Adversarial Constraints
    Dhruv Sarkar, Aprameyo Chakrabartty, Subhamon Supantha, Palash Dey, and Abhishek Sinha
    In The 29th International Conference on Artificial Intelligence and Statistics, 2026
  4. Relation-Aware Slicing in Cross-Domain Alignment
    Dhruv Sarkar, Aprameyo Chakrabartty, Anish Chakrabarty, and Swagatam Das
    In The 29th International Conference on Artificial Intelligence and Statistics, 2026
  5. Revisiting Social Welfare in Bandits: UCB is (Nearly) All You Need
    Dhruv Sarkar, Nishant Pandey, and Sayak Ray Chowdhury
    In The 29th International Conference on Artificial Intelligence and Statistics, 2026
  6. DP-NCB: Privacy Preserving Fair Bandits
    Dhruv Sarkar, Nishant Pandey, and Sayak Ray Chowdhury
    In The 40th AAAI Conference on Artificial Intelligence, 2026
  7. MATH4AI
    Optimal Anytime Algorithms for Online Convex Optimization with Adversarial Constraints
    Dhruv Sarkar, and Abhishek Sinha
    In AAAI 2026 Workshop on Foretell of Future AI from Mathematical Foundation, 2026
    Workshop publication; later accepted at ICML 2026.
  8. Nonlinear Two-Time-Scale Stochastic Approximation: A Sharp Phase Transition and How to Beat It
    Dhruv Sarkar, and Vaneet Aggarwal
    Preprint, 2026
  9. Improved Guarantees for Constrained Online Convex Optimization via Self-Contraction
    Dhruv Sarkar, and Abhishek Sinha
    Preprint, 2026
  10. A Simple Reduction Scheme for Constrained Contextual Bandits with Adversarial Contexts via Regression
    Dhruv Sarkar, and Abhishek Sinha
    Preprint, 2026
    Under submission at NeurIPS 2026.
  11. Online Learning for Approximately-Convex Functions with Long-Term Adversarial Constraints
    Dhruv Sarkar, Samrat Mukhopadhyay, and Abhishek Sinha
    Preprint, 2026
    Under submission at NeurIPS 2026.

2024

  1. Do We Really Need Foundation Models for Multi-Step-Ahead Epidemic Forecasting?
    Mrinmoy Dey*, Aprameyo Chakrabartty*Dhruv Sarkar*, and Tanujit Chakraborty
    In NeurIPS 2024 Workshop on Time Series in the Age of Large Models, 2024