My research focuses on the following key topics.
Game-theoretic Learning with Bandit Feedback
Problems in computer networks often arise from the interactions among multiple agents. In this research, we theoretically analyze scenarios involving multiple agents engaged in normal-form or extensive-form games with incomplete information and bandit feedback, employing uncoupled learning techniques. Specifically, we design bandit algorithms that guarantee convergence to equilibrium when adopted by all agents in the game, leveraging swap-regret-minimization techniques. Additionally, we explore the use of adaptive (optimistic or predictive) learning techniques to accelerate the convergence rate of these bandit algorithms.
Related Publications:
Faster Convergence for Unknown-game Bandits, Accepted by IEEE International Conference on Computer Communications (INFOCOM 2025)
Zhiming Huang, Jianping Pan
Game-theoretic Bandits for Network Optimization with High-probability Swap-regret Upper Bounds, 2024, IEEE/ACM Transactions on Networking (TON)
Zhiming Huang, Jianping Pan
A Near-optimal High-probability Swap-Regret Upper Bound for Multi-agent Bandits in Unknown General-sum Games, 2023, Uncertainty in Artificial Intelligence (UAI 2023)
Zhiming Huang, Jianping Pan
Combinatorial Learning with Bandit Feedback
Combinatorial optimization is also an important topic in computer networks. For example, a routing path selection involves a combination of feasible links. In this research, we are interested in the online optimization problem with bandit feedback, using Thompson sampling or Follow-the-regularized-learder (FTRL) techniques.
Related Publications:
Adversarial Semi-Bandits with Moving Arms, Accepted by IEEE International Conference on Computer Communications (INFOCOM 2025), May 2025
Zhiming Huang, Jianping Pan
Gaussian Randomized Exploration for Semi-bandits with Sleeping Arms, NeurIPS 2024 Workshop on Bayesian Decision-making and Uncertainty, Dec 2024.
Zhiming Huang, Bingshan Hu, Jianping Pan
Poster: Multi-agent Combinatorial Bandits with Moving Arms (Best Poster Award), IEEE International Conference on Distributed Computing Systems (ICDCS 2021)
Zhiming Huang, Bingshan Hu, Jianping Pan
Applications of Learning in Computer Networks
We apply the tools we studied above to concrete network problems, such as end-to-end congestion control, heterogenous network selection, caching and routing problems.
Related Publications:
Distributed Learning of Unknown Games for HetNet Selection, IEEE Transactions on Network Science and Engineering (TNSE)
Zhiming Huang, Jianping Pan
End-to-end Congestion Control as Learning for Unknown Games with Bandit Feedback, IEEE International Conference on Distributed Computing Systems (ICDCS 2023)
Zhiming Huang, Kaiyang Liu, Jianping Pan
TSOR: Thompson Sampling-based Opportunistic Routing, IEEE Transactions on Wireless Communications (TWC)
Zhiming Huang, Yifan Xu, Jianping Pan
Caching by User Preference with Delayed Feedback for Heterogeneous Cellular Networks, IEEE Transactions on Wireless Communications (TWC)
Zhiming Huang, Bingshan Hu, Jianping Pan