I am Bingshan Hu. I defended my thesis in Sep 2021 advised by Nishant Mehta at the University of Victoria. I am working on the theoretical side of machine learning, particularly, randomized learning algorithms for solving sequential decision making problems.
Before coming to Canada for my PhD study, I was a 3GPP delegate of SONY. I filed around 20 patents about how to use unlicensed spectrum from cellular operator's perspective and Device-to-Device communications. My full patent list can be found in Bingshan's Patents.
Google Scholar page: Bingshan's Google Scholar
Emails: bingshanhu3@gmail.com (preferred), bingsha1@cs.ubc.ca
Starting from Nov 2023 , I am a post-doc in UBC Data Science Institute co-hosted by Danica Sutherland and Mathias Lecuyer at the University of British Columbia
From Nov 2021 to Oct 2023, I was an Amii post-doc co-hosted by Nidhi Hegde at the University of Alberta, and Mark Schmidt and Mathias Lecuyer at the University of British Columbia
Education
Sep 2016 - Sep 2021: University of Victoria (UVic), Victoria, Canada
Ph.D. in Department of Computer Science
Thesis topic: Bandit algorithms with graphical feedback models and privacy awareness, nominated for Governor General’s Gold Medal by Department of Computer Science
Sep 2006 – Mar 2013: B.S. and M.S. in Communication Engineering, Beijing University of Posts and Telecommunications (BUPT), Beijing, China
Teaching
Co-instructor for statistical learning theory course in CS-UBC, topics including VC theory, Littlestone dimension, Hedge learning, and stochastic bandits, Lecture Notes
Publications and preprints
Efficient Kernelized Bandit Algorithms via Exploration Distributions, Bingshan Hu, Zheng He, and Danica Sutherland, 2025, PDF
Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret, Bingshan Hu, Zhiming Huang, Tianyue H. Zhang, Mathias Lecuyer and Nidhi Hegde, ICML 2025, PDF
Gaussian Randomized Exploration for Semi-bandits with Sleeping Arms, Zhiming Huang, Bingshan Hu, and Jianping Pan, Bayesian Decision-making and Uncertainty NeurIPS 2024 Workshop, 2024
Open Problem: Optimal Rates for Stochastic Decision-Theoretic Online Learning under Differential Privacy, Bingshan Hu and Nishant Mehta, COLT Open Problem, 2024, PDF
Efficient and Adaptive Posterior Sampling Algorithms for Bandits, Bingshan Hu, Zhiming Huang, Tianyue H. Zhang, Mathias Lécuyer, and Nidhi Hegde, 2024, PDF
(Near)-Optimal Algorithms For Differentially Private Online Learning in a Stochastic Environment, Bingshan Hu, Zhiming Huang, Nishant Mehta, and Nidhi Hegde, 2024, PDF
From 6235149080811616882909238708 to 29: Vanilla Thompson Sampling Revisited, Bingshan Hu and Tianyue H. Zhang, OPT NeurIPS 2023 Workshop, PDF
Differentially Private Algorithms for Efficient Online Matroid Optimization, Kushagra Chandak, Bingshan Hu, and Nidhi Hegde, CoLLAs 2023, PDF
Optimistic Thompson Sampling-based Algorithms for Episodic Reinforcement Learning, Bingshan Hu, Tianyue H. Zhang, Nidhi Hegde, and Mark Schmidt, UAI 2023, PDF
Near-Optimal Thompson Sampling-based Algorithms for Differentially Private Stochastic Bandits, Bingshan Hu and Nidhi Hegde, UAI 2022, PDF
Problem-Dependent Regret Bounds for Online Learning with Feedback Graphs, Bingshan Hu, Nishant Mehta, and Jianping Pan, UAI 2019, PDF
Poster: Multi-agent Combinatorial Bandits with Moving Arms, Zhiming Huang, Bingshan Hu, and Jianping Pan, IEEE ICDCS 2021, Best Poster Award
Caching by User Preference with Delayed Feedback for Heterogeneous Cellular Networks, Zhiming Huang, Bingshan Hu, and Jianping Pan, IEEE Transactions on Wireless Communications (TWC), 2020
Intelligent Caching Algorithms in Heterogeneous Wireless Networks with Uncertainty, Bingshan Hu, Yunjin Chen, Zhiming Huang, Nishant Mehta, and Jianping Pan, IEEE ICDCS 2019
Intelligent Caching in Dense Small-cell Networks with Limited External Resources, Bingshan Hu, Maryam Tanha, Dawood Sajjadi, and Jianping Pan, IEEE LCN 2018
Talks
Power of Data Randomization in Sequential Decision Making, University of British Columbia, AIM-SI Workshop: AI-Guided Experimental Design, Oct 2025
Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret, University of British Columbia, DSI Research Day, Jul 2025
Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret, Mila, The RL Sofa, Jul 2025
Efficient and Adaptive Thompson Sampling Algorithms for Bandits, Amii and University of Alberta, Amii Seminar, Jan 2025
Efficient and Adaptive Thompson Sampling Algorithms for Bandits, Simon Fraser University, VCR/AI Seminar, Dec 2024
Exploration-Driven Thompson Sampling in Reinforcement Learning, University of British Columbia, DSI Research Day, Jul 2024
(Near)-optimal Regret Bound for Differentially Private Thompson Sampling, Amii and University of Alberta, Amii Seminar, Jan 2022
Graphical Feedback Models for Sequential Decision Making, Amii and University of Alberta, interview talk for Amii Postdoctoral Fellowship, Apr 2021