Research map / Reinforcement learning / Imitation learning
Offline Imitation Learning with a Misspecified Simulator
Shengyi Jiang, Jingcheng Pang, Yang Yu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| match state distributions over an H-step horizon with adaptive inverse dynamics instead of one-step matching (GAILfO+ID) (ablation) | HIDIL · MuJoCo modified dynamics | normalized score: 64.8% -> 78.1% | Improved |
| replace PPO+ID simulator training with GAILfO-style state matching | inverse dynamics based sim-to-real imitation · MuJoCo modified dynamics | normalized score: 32.8% -> 64.8% | Improved |
| use HIDIL instead of behavior cloning on expert pairs | offline imitation with misspecified simulator · MuJoCo modified dynamics | normalized score: 55.0% -> 78.1% | Improved |
| vary horizon H (ablation) | HIDIL · MuJoCo modified dynamics | reward | No clear effect |
Closest papers by meaning
- Reward-free World Models for Online Imitation Learning (ICML 2025 · similarity 0.89)
- Hybrid Inverse Reinforcement Learning (ICML 2024 · similarity 0.89)
- Efficient Imitation under Misspecification (ICLR 2025 · similarity 0.89)
- Limited Preference Aided Imitation Learning from Imperfect Demonstrations (ICML 2024 · similarity 0.88)
- Imitation from Observations with Trajectory-Level Generative Embeddings (NeurIPS 2026 · similarity 0.88)
- Policy Decorator: Model-Agnostic Online Refinement for Large Policy Model (ICLR 2025 · similarity 0.88)
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.
