Yizhou Huang

Papers

Total Citations

H-Index

About

Yizhou Huang is a researcher at the forefront of reinforcement learning and multi-agent robotics, with a focus on creating adaptive, intelligent systems that operate effectively in dynamic environments. His most impactful work, "Continual Model-Based Reinforcement Learning with Hypernetworks" (2021), has garnered 20 citations and addresses a critical limitation in model-based reinforcement learning (MBRL) and model-predictive control (MPC): the assumption of stationary dynamics models. By introducing hypernetworks, Huang enables models to continually adapt to non-stationary environments without retraining from scratch, significantly improving planning accuracy and efficiency. This contribution is foundational for real-world applications where conditions change over time. Earlier, Huang explored decentralized coordination in "Learning of Coordination Policies for Robotic Swarms" (2017), tackling the challenge of designing scalable, distributed policies for large-scale robotic teams inspired by biological swarms. More recently, his work on "Stochastic Planning for ASV Navigation Using Satellite Images" (2023) extends these principles to autonomous surface vessels, using satellite imagery for robust navigation in uncertain aquatic environments. With a growing citation record and a clear trajectory toward addressing real-world deployment challenges, Huang’s research is shaping the future of continual learning and autonomous systems.