Max Robotics — Global Service Robot Directory + US Certification

About

Tianpei Yang is a prominent researcher specializing in deep reinforcement learning (DRL), multiagent reinforcement learning (MARL), and human-in-the-loop AI systems. His work addresses some of the most pressing challenges in the field, particularly the sample inefficiency that plagues modern RL agents across domains such as game AI, autonomous vehicles, and robotics. Yang's most influential contribution is his comprehensive survey on exploration in deep reinforcement learning, spanning both single-agent and multiagent settings, which has accumulated an impressive 158 citations and stands as a key reference for researchers navigating this complex landscape. His 2024 survey on Human-in-the-Loop Reinforcement Learning, already garnering 131 citations, reframes RL as an inherently human-centered paradigm, articulating the requirements, challenges, and opportunities that arise when humans remain integral to agent learning and deployment. Beyond surveys, Yang has made meaningful technical contributions through work on adaptive human reward shaping, graph neural network-based transfer learning across reinforcement learning tasks, and multi-task reinforcement learning with task-specific feature selection. His career arc—from early explorations of interactive human feedback in 2018 to today's systems-level thinking—reflects a consistent commitment to making reinforcement learning more efficient, transferable, and practically aligned with human needs.