Adish Singla

Max Planck Institute for Software Systems

Papers

Total Citations

H-Index

About

Adish Singla is a leading researcher at the intersection of machine learning, artificial intelligence, and human-computer interaction, with a core focus on the dynamics of teaching and learning between agents. His work fundamentally rethinks how AI systems can actively teach or interact with other learning agents—whether human or machine—rather than simply being trained on static data. In his highly cited 2018 paper, "Learning to Interact With Learning Agents," Singla laid the groundwork for understanding the unique challenges that arise when AI must engage with agents that are themselves learning and evolving. This work has been pivotal in shaping the field of interactive machine learning. More recently, his 2021 paper, "The Sample Complexity of Teaching by Reinforcement on Q-Learning," introduced a rigorous theoretical framework for "teaching dimension" in reinforcement learning, distinguishing between teaching via rewards versus demonstrations. This contribution provides a formal understanding of how efficiently a teacher can guide a student agent, with direct implications for robotics and personalized tutoring systems. With a growing citation footprint, Singla’s research is essential reading for anyone interested in multi-agent systems, algorithmic teaching, and the future of human-AI collaboration.