Ph.D. Computer Science and Engineering, The Chinese University of Hong Kong, 2020

B.E. Cryptography and Information Security, Shanghai Jiao Tong University, 2014 

Baoxiang Wang is an Assistant Professor at the School of Data Science, The Chinese University of Hong Kong, Shenzhen. Baoxiang graduated from Shanghai Jiao Tong University in 2014 with a Bachelor's degree in Cryptography and Information Security, and obtained his PhD degree in Computer Science and Engineering from The Chinese University of Hong Kong in 2020. He visited University of Alberta and Royal Bank of Canada for 16 months during his PhD.

Baoxiang Wang's research interests include reinforcement learning, online learning, and learning theory. His work are published at international conferences such as ITCS, NeurIPS, ICML, and ICLR. His work on The Gambler's problem answers the open question raised in 1998 in Sutton and Barto's reinforcement learning textbook, and mathematically proves the existence of chaos in reinforcement learning.

1. Baoxiang Wang, Shuai Li, Jiajin Li, Siu On Chan (2020). The Gambler's Problem and Beyond, International Conference on Learning Representations. 

2. Andrej Bogdanov, Baoxiang Wang (2020).  Learning and Testing Variable Partitions
Innovations in Theoretical Computer Science. 

3. Baoxiang Wang, Nidhi Hegde (2019). Privacy-preserving Q-Learning with Functional Noise in Continuous Spaces, Advances in Neural Information Processing Systems. 

4. Baoxiang Wang (2019). Recurrent Existence Determination Through Policy Optimization, International Joint Conference on Artificial Intelligence. 

5. Kenny Young, Baoxiang Wang, Matthew E. Taylor (2019). Metatrace Actor-Critic: Online Step-size Tuning by Meta-gradient Descent for Reinforcement Learning Control, International Joint Conference on Artificial Intelligence. 

6. Baoxiang Wang, Tongfang Sun, Xianjun Sam Zheng (2019). Beyond Winning and Losing: Modeling Human Motivations and Behaviors Using Inverse Reinforcement Learning, Artificial Intelligence and Interactive Digital Entertainment.

7. Jiajin Li, Baoxiang Wang (2018). Policy Optimization with Second-Order Advantage Information, International Joint Conference on Artificial Intelligence.

8. Shuai Li, Baoxiang Wang, Shengyu Zhang, Wei Chen (2016). Contextual Combinatorial Cascading Bandits, International Conference on Machine Learning.

9. Cuiyun Gao, Baoxiang Wang, Pinjia He, Jieming Zhu, Yangfan Zhou, Michael R. Lyu (2015). PAID: Prioritizing App Issues for Developers by Tracking User Reviews Over Versions, International Symposium on Software Reliability Engineering.