DualCOIL: Offline Imitation Learning from Contrasting Demonstrations
Speaker: 
HOANG Minh Huy
Ph.D. Candidate
School of Computing and Information Systems
Singapore Management University
| Date: Time: Venue: | | 30 June 2026, Tuesday 3:00pm – 3:30pm Meeting room 4.4, Level 4.
School of Computing and Information Systems 1,
Singapore Management University,
80 Stamford Road
Singapore 178902
Please register by 29 June 2026. 
|
|
About the Talk
Offline imitation learning typically learns from expert and unlabeled demonstrations, yet often overlooks the valuable signal in explicitly undesirable behaviors. In this work, we study offline imitation learning from contrasting behaviors, where the dataset contains both expert and undesirable demonstrations along with an unlabeled set of demonstrations. We propose a novel formulation that optimizes a difference of KL divergences over the state-action visitation distributions of expert and undesirable (or bad) data. Although the resulting objective is a DC (Difference-of-Convex) program, we prove that it becomes convex when expert demonstrations outweigh undesirable demonstrations, enabling a practical and stable non-adversarial training objective. Our method avoids adversarial training and handles both positive and negative demonstrations in a unified framework. Extensive experiments on standard offline imitation learning benchmarks demonstrate that our approach consistently outperforms state-of-the-art baselines.
This is a Pre-Conference talk for Forty-Third International Conference on Machine Learning (ICML 2026).
About the Speaker
HOANG Minh Huy is a PhD candidate in Computer Science at Singapore Management University, under the supervision of Assistant Professor Tien Mai and Dr. Pavitra Krishnaswamy. His research focuses on Deep Reinforcement Learning and Imitation Learning, particularly developing novel algorithms for Offline Imitation Learning from mixed-quality data (including suboptimal and undesirable demonstrations) and Safe/Constrained Reinforcement Learning.
