About me

I am a researcher in Wireless Systems Laboratory (WSL) at National Institute of Information and Communication Technology (NICT), Japan. My current research focus on bringing formal methods from software engineering to analyze, verify, and improve the robustness of AI-enabled wireless systems.

News

• 9/2026: NeurIPS’26 (co-first) paper on applying neural combinatorial optimization to accelerate branching decision in NeuralSAT
• 9/2026: WPMC'26 paper, a software-defined radio implementation of semantic communication
• 7/2026: VTC Fall'26 paper, a verified DNN-based power allocation
• 2/2026: CVPR’26 highlight paper (co-author) on designing input robustness properties for image classification
• 1/2026: FSE’26 paper (co-first) on verifying signal processing-inspired robustness properties

Publications

1. “Verifying Neural Networks with Reinforcement Learning”
   Hai Duong*, Thanh Le*, ThanhVu Nguyen
   accepted at Neural Information Processing Systems (NeurIPS'26)
   PDF
2. “Implementation of Zero-shot Semantic Communication on Software Defined Radio”
   Thanh Le, Arif Dataesatu, Homare Murakami, Takeshi Matsumura
   accepted at Wireless Personal Multimedia Communications (WPMC'26)
   PDF Code
3. “Formal Verification for Deep Learning-based Power Control in Massive MIMO”
   Thanh Le, Takeshi Matsumura, Yusheng Ji, John C.S. Lui
   Vehicular Technology Conference (VTC Fall'26), Sep. 2026
   PDF Code
4. “Verifying Neural Network Robustness with Dual Perturbations”
   Hai Duong, Lam Nguyen, Thanh Le, ThanhVu Nguyen
   Computer Vision and Pattern Recognition (CVPR'26), June 2026, ☆Highlight☆
   PDF Code
5. “Verifying Structural Robustness of Deep Neural Network”
   Hai Duong*, Thanh Le*, Lam Nguyen, ThanhVu Nguyen
   Foundations of Software Engineering (FSE'26), July 2026
   PDF Code
6. “Formal Verification for Deep Learning-based Mobile Network Traffic Prediction”
   Thanh Le, Takeshi Matsumura
   International Conference on Artificial Intelligent in Information and Communication (ICAIIC'26), Feb. 2026
   PDF Code

Preprints

1. “Neural Network Verification for Deep Joint Source-Channel Coding”
   Thanh Le, Hai Duong, Takeshi Matsumura, ThanhVu Nguyen
   Submitted to IEEE Transactions on Machine Learning in Communications and Networking (TMLCN)
   PDF Code Gist
2. “Towards Verifying Neural Networks against Multi-Parameter Bit-Flip Perturbations”
   Hai Duong, Thanh Le, Ho Nguyen, ThanhVu Nguyen
   PDF
3. “CLAD: Constrained Abstract Domain for Neural Network Verification”
   Hai Duong, Thanh Le, ThanhVu Nguyen
   PDF
4. “ZonoGPT: Towards an Abstract Domain for Verifying Large GPT Models”
   Hai Duong, Thanh Le, ThanhVu Nguyen
   PDF
5. “FGGM: Formal Grey-box Gradient Method for Attacking DRL-based MU-MIMO Scheduler”
   Thanh Le, Hai Duong, Yusheng Ji, ThanhVu Nguyen, John C.S. Lui
   PDF

Dissertation

Distributed Learning Techniques for Dynamic Access Control in massive Machine-Type Communication
Supervisor: Prof. Yusheng Ji
SOKENDAI, NII, Mar. 2025
Abstract

1. “TinyQMIX: Distributed Access Control for mMTC via Multi-agent Reinforcement Learning”
   Thanh Le, Yusheng Ji, John C.S. Lui
   Vehicular Technology Conference (VTC Fall'22), Sep. 2022
   PDF
2. “MFTTS: A Mean-Field Transfer Thompson Sampling Approach for Distributed Power Allocation in Unsourced Multiple Access”
   Thanh Le, Yusheng Ji, John C.S. Lui
   Transactions on Mobile Computing (Volume: 23, Issue: 12, Dec. 2024)
   PDF

Services

• Review papers at ICC'22,23,25, GlobeCom'24,25, SMARTCOMP'25, TCCN, ComCom, LCN'26, WPMC'26 Wksp, ICLR'27.
• Help review papers at ICML'26, NeurIPS'26, AAAI'27, POPL'27.
• TPC member at INFOCOM'26 Wksp.

External Links

• Google Scholar
• Research Map
• Openreview
• Github
• ORCID
• DBLP

Tool & Software

1. NeuralSAT - A neural network verification tool
2. Abstract Domain Explainer - An interactive, step-by-step walkthrough of how DNN abstract domains work.
3. VietProfs - A directory of Vietnamese academic and research worldwide
4. CS Picks - CS Confs - Track upcoming CS conferences
5. CS Picks - Jobs - Track US academic job openings
6. An website (AI slop) to track performance of Vietnamese University in top AI conferences