Research map / Reinforcement learning / Imitation learning
Mitigating Covariate Shift in Imitation Learning via Offline Data With Partial Coverage
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add a pessimistic uncertainty penalty to the imitation objective (ablation) | MILO · Hopper-v2 and Walker2d-v2 | final performance | Helped |
| use lower-coverage offline data collected by a weaker behavior policy (ablation) | MILO · MuJoCo continuous control tasks | expert performance normalized score | Mixed |
| increase the expert demonstrations to one trajectory (ablation) | MILO · Walker2d-v2 and Humanoid-v2 | performance relative to expert | Helped |
| use MILO with offline transition data rather than behavior cloning on expert samples alone | offline imitation learning · five MuJoCo environments | performance | Helped |
| combine offline and expert samples with behavior cloning | BC · MuJoCo continuous control tasks | imitation performance | Mixed |
Closest papers by meaning
- Mitigating Covariate Shift in Behavioral Cloning via Robust Stationary Distribution Correction (NeurIPS 2024 · similarity 0.95)
- How to Leverage Diverse Demonstrations in Offline Imitation Learning (ICML 2024 · similarity 0.92)
- Is Behavior Cloning All You Need? Understanding Horizon in Imitation Learning (NeurIPS 2024 · similarity 0.92)
- OLLIE: Imitation Learning from Offline Pretraining to Online Finetuning (ICML 2024 · similarity 0.91)
- Behavior Cloning is Not All You Need: The Optimality of On-Policy Distillation for Noisy Expert Feedback (NeurIPS 2026 · similarity 0.91)
- Interactive and Hybrid Imitation Learning: Provably Beating Behavior Cloning (NeurIPS 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.