Thanh Nguyen-Tang

About Me

I am a Postdoctoral Research Fellow in Department of Computer Science at Johns Hopkins University, working with Raman Arora. I was an Associate Research Fellow at the Applied AI Institute, Deakin University in July 2021-June 2022 and completed my PhD there in Feb 2022. I did my Master in Computer Science and Engineering at Ulsan National Institute of Science and Technology (UNIST) in 2018.

Research Interests

I am building toward data-efficient, runtime-efficient and robust AI by studying three foundational pillars of modern machine learning – provable statistical efficiency, computational efficiency, and robustness. My current key focus includes:

• Reinforcement Learning

• (Deep) learning theory

Besides, I am also studying algorithmic designs for

• Learning under Distributional Shifts

• Robust Adversarial Learning

• Probabilistic Deep Learning

• Representation Learning

I’m always actively open to research collaborations and chat!

Recent News

• Jan. 20, 2023: One paper accepted to ICLR’23 (Spotlight, acceptance rate: 31.8%)

• Dec. 9, 2022: One paper accepted to TMLR.

• Nov. 19, 2022: One paper accepted to AAAI, 2023 (acceptance rate: 19.6%).

• Oct. 30, 2022: I was acknowledged in Francis Bach's “Learning Theory from First Principles”

• Sep. 14, 2022: One paper accepted to NeurIPS, 2022 (acceptance rate: 25.6%).

• Aug. 8, 2022: I am acknowledged in Mengyan Zhang's PhD thesis.

• Jan. 21, 2022: One paper accepted to ICLR, 2022 (acceptance rate: 32.26%).

• May 20, 2021: Accepted to the Deep Learning Theory Summer School at Princeton (acceptance rate: 180/500 = 36%).

Publications

2023

• Provably Efficient Neural Offline Reinforcement Learning via Perturbed Rewards
  

  • Thanh Nguyen-Tang, Raman Arora
    

  • International Conference on Learning Representations (ICLR), 2023 (Top 25% - Spotlight)
    

  • [slides]

• On Instance-Dependent Bounds for Offline Reinforcement Learning with Linear Function Approximation
  

  • Thanh Nguyen-Tang, Ming Yin, Sunil Gupta, Svetha Venkatesh, Raman Arora
    

  • AAAI Conference on Artificial Intelligence (AAAI), 2023
    

  • [arXiv] [poster] [slides] [video]

2022

• On Sample Complexity of Offline Reinforcement Learning with Deep ReLU Networks in Besov Spaces
  

  • Thanh Nguyen-Tang, Sunil Gupta, Hung Tran-The, Svetha Venkatesh
    

  • Transactions on Machine Learning Research (TMLR), 2022

  • [cite]

• TIPI: Test Time Adaptation with Transformation Invariance
  

  • A. Tuan Nguyen, Thanh Nguyen-Tang, Ser-Nam Lim, Philip Torr
    

  • Under review, 2022
    

• Improving Domain Generalization with Interpolation Robustness
  

  • Ragja Palakkadavath, Thanh Nguyen-Tang, Sunil Gupta, Svetha Venkatesh
    

  • Distribution Shifts Workshop@NeurIPS2022, INTERPOLATE@NeurIPS2022 (Spotlight)
    

  • Under review, 2022
    

• Learning Fractional White Noises in Neural Stochastic Differential Equations
  

  • Anh Tong, Thanh Nguyen-Tang, Toan Tran, Jaesik Choi
    

  • Advances in Neural Information Processing Systems (NeurIPS), 2022
    

  • [cite] [code]

• Two-Stage Neural Contextual Bandits for Adaptive Personalised Recommendation
  

  • Mengyan Zhang, Thanh Nguyen-Tang, Fangzhao Wu, Zhenyu He, Xing Xie, Cheng Soon Ong
    

  • Under review, 2022
    

  • [arXiv]

• Contextual Bandits with Reduced Explorations via Logged Data
  

  • Hung Tran-The, Thanh Nguyen-Tang, Sunil Gupta, Santu Rana, Svetha Venkatesh
    

  • Under review, 2022
    

  • [arXiv]

• Offline Neural Contextual Bandits: Pessimism, Optimization and Generalization
  

  • Thanh Nguyen-Tang , Sunil Gupta, A.Tuan Nguyen, and Svetha Venkatesh
    

  • International Conference on Learning Representations (ICLR), 2022
    

  • Workshop on OfflineRL, NeurIPS, 2021
    

  • [cite] [arXiv] [poster] [slides] [code]

2021

• Sample Complexity of Offline Reinforcement Learning with Deep ReLU Networks
  

  • Thanh Nguyen-Tang, Sunil Gupta, Hung Tran-The, Svetha Venkatesh
    

  • Workshop on Reinforcement Learning Theory, ICML, 2021
    

  • [arXiv] [slides] [talk]

• Distributional Reinforcement Learning via Moment Matching
  

  • Thanh Nguyen-Tang, Sunil Gupta, Svetha Venkatesh
    

  • AAAI Conference on Artificial Intelligence (AAAI), 2021
    

  • [cite] [arXiv] [code] [slides] [poster] [talk]

2020

• Distributionally Robust Bayesian Quadrature Optimization
  

  • Thanh Nguyen-Tang, Sunil Gupta, Huong Ha, Santu Rana, Svetha Venkatesh
    

  • International Conference on Artificial Intelligence and Statistics (AISTATS), 2020
    

  • [cite] [arXiv] [code] [slides] [talk]

2019

• Markov Information Bottleneck to Improve Information Flow in Stochastic Neural Networks
  

  • Thanh Nguyen-Tang, Jaesik Choi
    

  • Entropy, 21(10), 976, 2019
    

  • [cite] [code]

• Bayesian Optimization with Unknown Search Space
  

  • Huong Ha, Santu Rana, Sunil Gupta, Thanh Nguyen-Tang, Hung Tran-The, Svetha Venkatesh
    

  • Advances in Neural Information Processing Systems (NeurIPS), 2019
    

  • [cite] [code] [poster]

Dissertations

• On Practical Reinforcement Learning: Provable Robustness, Scalability and Statistical Efficiency
  

  • Ph.D. dissertation, Deakin University, Australia, July 2021

• Parametric Information Bottleneck to Optimize Stochastic Neural Networks
  

  • Master Thesis, Ulsan National University of Science and Technology, South Korea, 2018
    

  • [slides] [talk]

Academic Service

• Senior Program Committee: AAAI (2023)

• Reviewer/Program Committee:

  • NeurIPS (2022, 2021, 2020)

  • ICML (2023, 2022, 2021)

  • ICLR (2023, 2022, 2021- Outstanding reviewer award)

  • AAAI (2022, 2021-Top 25% PC, 2020)

  • TPAMI (2023)

  • AISTATS (2021)

  • EWRL (2022)

  • L4DC (2022)

  • NeurIPS Workshop on OfflineRL (2022, 2021)

• Volunteer: ICML (2022), AutoML (2022)

Invited Talks

• Neural Offline Reinforcement Learning [post]

  • VinAI, Vietnam, Jan. 13, 2023

• Neural Offline Reinforcement Learning [record]

  • FPT AI, Vietnam, Dec. 21, 2022

• Neural Offline Reinforcement Learning [slides]

  • UC San Diego, USA, Dec. 8, 2022 (Host: Rose Yu)

• Offline Reinforcement Learning: Assurance in High-Stakes AI Applications [slides]

  • IAA Research Summit, Johns Hopkins University, USA, Nov. 2022

• Offline Neural Contextual Bandits: Pessimism, Optimization, and Generalization

  • Ohio State University, USA, Jan. 2022 (Host: Yingbin Liang and Ness Shroff)

• Offline Neural Contextual Bandits: Pessimism, Optimization, and Generalization

  • Arizona State University, USA, Dec. 2021 (Host: Kwang-Sung Jun)

• Generalization and Optimization in Deep Learning: Over-parameterization and Interpolation [slides]

  • Deakin University, Australia, Aug. 2021

• On Finite-Sample Analysis of Batch Reinforcement Learning with Deep ReLU Networks

  • Viet Operator Theorists Group, Vietnam and USA, Apr. 2021