Nitin Ragothaman

About

Nitin Ragothaman is a researcher advancing the frontiers of reinforcement learning (RL) in distributed and data-constrained settings. His primary research areas include federated learning, offline RL, and ensemble-based decision-making. In his notable 2023 work, "Federated Ensemble-Directed Offline Reinforcement Learning," Ragothaman tackles a critical challenge: enabling multiple distributed agents to collaboratively learn a robust control policy from small, heterogeneous pre-collected datasets without direct interaction. He proposes a novel framework that uses ensemble methods to guide policy learning, effectively addressing the pitfalls of naïvely combining decentralized data, such as distribution shift and conflicting behavior policies. This contribution is particularly impactful for privacy-sensitive or communication-limited applications like healthcare and autonomous systems. With 2 citations already, this paper is gaining recognition for its practical relevance. Ragothaman’s work stands out for its elegant synthesis of federated and offline RL paradigms, offering a scalable solution to real-world multi-agent learning problems. His research continues to influence how we think about cooperative, data-efficient decision-making under uncertainty.

Research Focus

Key Achievements

Top Papers

1. 1
   Federated Ensemble-Directed Offline Reinforcement Learning

   2 citations · 2023

Key Collaborators

Contact & Links