Sai Rajeswar
Papers
1
Total Citations
2
H-Index
1
About
Sai Rajeswar is a researcher advancing the frontiers of reinforcement learning and robotics, with a focus on enabling agents to learn efficiently in complex, contact-rich environments. His key research areas span curiosity-driven exploration, tactile sensing, and sparse-reward tasks—domains where traditional reinforcement learning often struggles. Rajeswar’s most notable contribution is his work on "Touch-based Curiosity for Sparse-Reward Tasks" (2021), which introduces a novel framework that leverages surprise from mismatches in tactile feedback to guide exploration. By integrating force/torque sensors commonly found in robotic grippers, his approach allows robots to autonomously discover meaningful behaviors in tasks requiring fine-grained physical interaction, such as assembly or manipulation. This work, cited 2 times, represents a foundational step toward more sample-efficient and physically aware learning systems. Rajeswar’s research bridges the gap between curiosity-driven AI and real-world robotic applications, offering a path to reduce reliance on handcrafted rewards. His contributions are particularly valuable for students and researchers interested in embodied AI, reinforcement learning, and the intersection of perception and action in robotics.
Research Focus
Key Achievements
Top Papers
- 1Touch-based Curiosity for Sparse-Reward Tasks2 citations · 2021