Yuhu Cheng

China University of Mining and Technology, Xuzhou University of Technology

Papers

Total Citations

117

H-Index

About

Yuhu Cheng is a researcher at the forefront of artificial intelligence, with key contributions spanning deep reinforcement learning, emotion recognition, and robotics. His work is distinguished by a dual focus on both advancing theoretical foundations and developing practical algorithms. In reinforcement learning, Cheng introduced the **Authentic Boundary Proximal Policy Optimization** (2021, 59 citations), which provides a novel theoretical explanation for the clipping mechanism in PPO, a widely-used algorithm. He further advanced the field with **Off-Policy Deep Reinforcement Learning Based on Steffensen Value Iteration** (2020, 15 citations), addressing the sample inefficiency inherent in trial-and-error learning. In affective computing, his **Dynamic Complementary Graph Convolutional Network** (2024, 31 citations) tackles the challenging task of emotion recognition in conversation by efficiently modeling contextual information. Earlier in his career, Cheng also made notable contributions to robotics with an **Improved ant colony algorithm for mobile robot path planning** (2012, 12 citations), which enhanced path smoothness and convergence. Across these diverse domains, Cheng’s research consistently demonstrates a commitment to solving fundamental algorithmic limitations, making his work highly influential for students and researchers seeking to push the boundaries of intelligent systems.