Research

Theory and methods in high-dimensional statistics, robust and stochastic optimization, reinforcement learning, language and diffusion models, sampling, and differential privacy.

2026
Provably Reliable Classifier Guidance through Cross-entropy Error Control
Sharan Sahu, Arisina Banerjee, and Yuchen Wu
The 40th Annual Conference on Neural Information Processing Systems (NeurIPS 2026)
arXiv
On the Provable Suboptimality of Momentum SGD in Nonstationary Stochastic Optimization
Sharan Sahu, Cameron J. Hogan, and Martin T. Wells
The 43rd International Conference on Machine Learning (ICML 2026)
arXiv | ICML Poster
Adapt or Forget: Provable Tradeoffs Between Adam and SGD in Nonstationary Optimization
Sharan Sahu, Abir Sarkar, Cameron J. Hogan, and Martin T. Wells
Preprint, 2026
arXiv
HOPPER: Learnable Hop Extraction for Linearized Graph Sequence Models
Isuru Herath, Arin Gopakumar, and Sharan Sahu
Preprint, 2026
arXiv
2025
Online Distributionally Robust LLM Alignment via Regression to Relative Reward
Sharan Sahu and Martin T. Wells
Preprint, 2025
arXiv
Towards Optimal Differentially Private Regret Bounds in Linear Markov Decision Processes
Sharan Sahu
Preprint, 2025
arXiv