Ruosong Wang

Ruosong Wang is an Assistant Professor at Peking University, focusing on Reinforcement Learning, Neural Networks, and Optimization. His research bridges theoretical analysis with practical implementations in AI, particularly in deep learning and reinforcement learning frameworks.

Research Interests: Wang’s work explores the intersection of Machine Learning Theory, Function Approximation, Generalization, and Algorithm Design. He investigates how overparameterized neural networks and infinitely wide deep nets can be optimized for small-data tasks, while addressing challenges in offline RL, preference-based learning, and graph neural networks.

Article Trends: His publications (2019–2021) emphasize provably efficient algorithms for reinforcement learning, fine-grained theoretical analysis of neural network optimization, and hybrid models combining graph kernels with neural architectures. Key themes include sample efficiency, representation learning, and bounded eluder dimension in RL.

Collaborations and Access: Wang collaborates with leading researchers at institutions like Carnegie Mellon University and UC Berkeley. All his articles are publicly accessible through arXiv or conference proceedings, with citations reflecting significant influence in machine learning and AI communities.