Thang Duong

About

Selected Research

Published NeurIPS 2024

Beyond Task Diversity: Provable Representation Transfer for Sequential Multi-Task Linear Bandits

Thang Duong, Zhi Wang, Chicheng Zhang

Established a regret guarantee for sequential representation transfer without the standard task-diversity assumption, showing that this assumption is not necessary for the guarantee.

I proposed the exponential-algorithm approach after earlier proof attempts based on Davis-Kahan could not remove task diversity. BOSS learns and transfers a low-rank representation across tasks under an ellipsoid action-set assumption. Its computational cost motivates our follow-up work on efficient methods.

PDF arXiv Code

BibTeX

@article{duong2024beyond,
  title   = {Beyond task diversity: provable representation transfer for sequential multitask linear bandits},
  author  = {Duong, Thang and Wang, Zhi and Zhang, Chicheng},
  journal = {Advances in Neural Information Processing Systems},
  volume  = {37},
  pages   = {37791--37822},
  year    = {2024}
}

Copy

Published WiOpt 2026

Physics-Informed Parametric Bandits for Beam Alignment in mmWave Communications

Hao Qin*, Thang Duong*, Ming F. Li, Chicheng Zhang

2× reduction in beam-alignment regret on the real-world DeepMIMO and DeepSense6G benchmarks, including settings where unimodality or multimodality assumptions break.

We cast mmWave beam alignment as a parametric bandit tied to the Phase-Retrieval structure of the channel and design two algorithms (PR-ETC, PR-GREEDY) that exploit sparse multipath structure without restrictive assumptions on the reward function.

arXiv

Published ICML 2025 EXAIT Workshop

Improving the Data-efficiency of Reinforcement Learning by Warm-starting with LLM

Thang Duong, Minglai Yang, Chicheng Zhang

Studied whether LLM-collected trajectories can improve the early learning of RL agents in small Gymnasium environments.

LORO uses LLM-generated data to warm-start an RL agent before online training. The workshop paper evaluates the approach on classical MDP benchmarks; the linked repository also contains subsequent experiments.

Workshop arXiv Code

Working paper

Efficient Algorithms for Lifelong Representation Learning in Linear Bandits Beyond Task Diversity

Thang Duong, et al.

A computationally efficient algorithm (BRESS) that drops the task-diversity assumption and remains polynomial-time via a black-box reduction to online PCA, with adversarial robustness.

By reducing sequential multi-task linear bandits to online principal component analysis, any low-competitive-ratio online-PCA subroutine yields low meta-regret. (Details available upon publication.)

Published RLC 2024

Non-stationary Bandits and Meta-Learning with a Small Set of Optimal Arms

MJ Azizi, Thang Duong, Yasin Abbasi-Yadkori, András György, Claire Vernade, M. Ghavamzadeh

Meta-learns across a sequence of bandit tasks whose optimal arms lie in a small unknown subset, beating the naive Õ(√(KNT)) baseline.

Via a reduction to bandit submodular maximization, the algorithm exploits shared optimal-arm structure across tasks in both the meta-learning and non-stationary settings.

PDF arXiv Code

Interactive: bandit regret

A core theme of my work is sample efficiency: how fast a learner stops paying for exploration. Below, two classic strategies play the same multi-armed bandit live in your browser: UCB (adaptive, logarithmic regret) vs. Explore-Then-Commit (explores, then commits). Drag the sliders to see how the gap between arms and the horizon change the cost of learning. (Illustration simulated here, not paper data.)

Publications

1. Beyond Task Diversity: Provable Representation Transfer for Sequential Multi-Task Linear Bandits

   Thang Duong, Zhi Wang, Chicheng Zhang

   NeurIPS , 2024 [PDF] [arXiv] [Code]

2. Physics-Informed Parametric Bandits for Beam Alignment in mmWave Communications

   Hao Qin*, Thang Duong*, Ming F. Li, Chicheng Zhang

   WiOpt , 2026 [arXiv]

3. Improving the Data-efficiency of Reinforcement Learning by Warm-starting with LLM

   Thang Duong, Minglai Yang, Chicheng Zhang

   ICML EXAIT Workshop , 2025 [Workshop] [arXiv] [Code]

4. Efficient Algorithms for Lifelong Representation Learning in Linear Bandits Beyond Task Diversity

   Thang Duong, et al.

   Working paper

5. Non-stationary Bandits and Meta-Learning with a Small Set of Optimal Arms

   MJ Azizi, Thang Duong, Yasin Abbasi-Yadkori, András György, Claire Vernade, M. Ghavamzadeh

   RLC , 2024 [PDF] [arXiv] [Code]

6. Association of MRI-defined Structure Features at Baseline with Knee Pain Trajectories

   S. Liu, X. Sun, Y. Ge, Thang Duong, C.K. Kwoh

   ACR Convergence , 2024 [Link]

7. Deep Regression for Precise Geometric Dimension Measurement

   Thang Duong, Binh Nguyen Duc, Phuong Le Khac, Ngoc Tu Nguyen, Mai Nguyen Thi Phuong

   J. Korean Soc. Precis. Eng. , 2019 [PDF] [DOI]

8. Analyzing Seismic Signal using Support Vector Machine for Vehicle Motion Detection

   Thang Duong, Nguyen Thi Phuong Mai

   J. Industrial Networks and Intelligent Systems , 2019 [PDF]

Experience

Education

Skills