Malayandi Palan

Papers

Total Citations

H-Index

About

Malayandi Palani is a robotics and artificial intelligence researcher specializing in reward learning, human-robot interaction, and preference-based learning. His work addresses one of the central challenges in autonomous robotics: how machines can accurately infer human objectives without requiring manually engineered reward functions. Palani's most influential contribution, "Asking Easy Questions: A User-Friendly Approach to Active Reward Learning" (2019, 54 citations), introduced a human-centric framework for active reward learning that prioritizes query accessibility alongside information gain — a meaningful departure from prior methods that ignored cognitive burden on human teachers. This work reflects his broader commitment to making robot learning practical and inclusive for real-world users. His subsequent research deepened this foundation by exploring how robots can optimally integrate multiple streams of human feedback. "Learning Reward Functions from Diverse Sources of Human Feedback" (2021, 21 citations) and its earlier companion paper (2019, 18 citations) demonstrated how combining passive demonstrations with active preference queries yields more efficient and accurate reward models than either approach alone. Collectively, Palani's research has meaningfully advanced the field of inverse reward design, offering frameworks that are both theoretically grounded and sensitive to the practical realities of human-robot collaboration. His work is essential reading for researchers in robot learning and human-AI alignment.