Yiming Zhang

National University of Defense Technology, Supélec

Papers

Total Citations

329

H-Index

About

Yiming Zhang is a prominent artificial intelligence researcher whose work sits at the intersection of deep learning, reinforcement learning, and safe autonomous decision-making. His most influential contribution, the comprehensive survey "Deep Reinforcement Learning: A Survey" (2020), has garnered an impressive 277 citations, reflecting its widespread adoption as a foundational reference for researchers and practitioners entering the field. The paper systematically examines deep RL's evolution and its far-reaching applications across domains including robotic control, end-to-end autonomous systems, recommendation engines, and natural language dialogue systems. Beyond surveying the landscape, Zhang has made meaningful theoretical contributions to constrained optimization in reinforcement learning. His work "First Order Constrained Optimization in Policy Space" (2020), with 52 citations, addresses a critical challenge in deploying RL agents safely — ensuring that learned behaviors not only maximize reward but also respect safety constraints that prevent harmful or undesirable actions. This dual focus on broad synthesis and rigorous algorithmic development positions Zhang as a researcher who both maps the terrain of modern AI and actively advances its frontiers, particularly in building reliable, safe, and high-performing autonomous systems.