He Ba

About

He Ba is a researcher advancing the frontier of reinforcement learning and continuous control, with a focus on bridging planning and deep actor-critic methods. His most-cited work, "Critic PI2," introduces a novel framework that integrates Policy Improvement with Path Integrals into deep reinforcement learning, enabling agents to master continuous planning tasks—a domain where traditional tree-based methods like AlphaGo and MuZero excel in discrete settings but struggle. This contribution addresses a critical gap in AI, offering a pathway for planning in real-world, continuous environments such as robotics and autonomous systems. Though early in its citation impact (2 citations), the work signals a promising direction for scalable, model-based RL. Ba’s research underscores a commitment to constructing agents with robust planning capabilities, tackling challenges that span from theoretical foundations to practical deployment. His efforts contribute to the broader pursuit of artificial intelligence that can reason and act in complex, continuous spaces, making his work of particular interest to students and researchers exploring the next generation of reinforcement learning algorithms.

Research Focus

Key Achievements

Top Papers

1. 1
   Critic PI2: Master Continuous Planning via Policy Improvement with Path Integrals and Deep Actor-Critic Reinforcement Learning

   2 citations · 2021

Key Collaborators

Contact & Links