Max Robotics — Global Service Robot Directory + US Certification

About

Isaac Sheidlower is a researcher at the forefront of human-robot interaction, specializing in interactive reinforcement learning (IntRL) and human-in-the-loop robotics. His work addresses a critical challenge: how to make autonomous robots not just intelligent, but also teachable and controllable by everyday users. Sheidlower’s major contributions include pioneering methods for human teachers to guide robot learning in complex, continuous action spaces—a significant leap beyond traditional discrete-action IntRL, as demonstrated in his most-cited paper (8 citations). He has also systematically studied how robot errors shape human teaching behavior, revealing the dynamic, adaptive nature of human instruction. A standout achievement is his development of "imagined actions" and "in-distribution states" frameworks, which empower users to creatively repurpose a robot’s learned policies for novel tasks by leveraging predictable behavior. With a growing citation impact and a clear focus on user agency, Sheidlower’s research bridges the gap between advanced reinforcement learning algorithms and practical, human-centered robot deployment, making him a key voice in the future of collaborative robotics.