Koustubh Sharma
Papers
1
Total Citations
46
H-Index
1
About
Koustubh Sharma is a researcher at the forefront of reinforcement learning and energy-based modeling, with a focus on developing principled frameworks for intelligent decision-making. His most cited work, "Maximum Information Measure Policies in Reinforcement Learning with Deep Energy-Based Model" (2021, 46 citations), introduced a novel approach to acquiring articulated electricity regulations for consistent states and actions. This contribution provided a theoretical foundation for learning maximum entropy policies, which has since been adapted by developers to create a simplified Q-learning service, bridging the gap between complex energy-based models and practical reinforcement learning applications. Sharma's research addresses the longstanding challenge of making maximum entropy policy acquisition attainable beyond summarised domains, offering a scalable solution that has influenced subsequent work in deep reinforcement learning. His efforts have helped advance the understanding of how information-theoretic measures can guide policy optimization, making his work a valuable reference for students and researchers exploring the intersection of energy-based models and reinforcement learning.
Research Focus
Key Achievements
Top Papers
- 1