Max Robotics — Global Service Robot Directory + US Certification

About

Te Sun is a leading researcher in continual and demonstration-guided reinforcement learning (RL), with a focus on enabling robots to learn and adapt across multiple tasks without catastrophic forgetting. His pioneering work on policy distillation and sim-to-real transfer, most notably in "DisCoRL: Continual Reinforcement Learning via Policy Distillation" (2019, 35 citations), introduced a framework that allows a single model to sequentially learn and retain distinct policies, addressing both training-time multi-task learning and test-time task inference without external signals. Sun extended this to real-world deployment in "Continual Reinforcement Learning deployed in Real-life using Policy Distillation and Sim2Real Transfer" (2019, 17 citations), demonstrating a robot’s ability to solve sequentially presented tasks while preserving past knowledge. He further advanced sample efficiency in sparse-reward environments through demonstration-guided approaches, such as "Exploration-efficient Deep Reinforcement Learning with Demonstration Guidance for Robot Control" (2020, 6 citations) and "Demonstration Guided Actor-Critic Deep Reinforcement Learning for Fast Teaching of Robots in Dynamic Environments" (2020, 4 citations). By integrating expert demonstrations into actor-critic architectures, Sun’s work significantly reduces interaction costs and stabilizes training, offering practical solutions for deploying RL in dynamic, real-world robotic settings.