Research map / Reinforcement learning / Imitation learning
Confidence-Aware Imitation Learning from Demonstrations with Varying Optimality
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 |
|---|---|---|---|
| use CAIL instead of 2IWIL (closest baseline) | CAIL · Reacher | expected return: -23.055 -> -7.816 | Helped |
| use CAIL instead of 2IWIL | CAIL · Ant | expected return: 3473.852 -> 3825.644 | Helped |
| use CAIL instead of 2IWIL | CAIL · Simulated Panda arm | expected return: -120.622 -> -62.946 | Helped |
| use CAIL instead of 2IWIL/AIRL | CAIL · Real UR5e arm | expected return: -52.445 -> -34.330 | Helped |
| use CAIL instead of 2IWIL on success rate | CAIL · Real robot arm | success rate (%): 83 -> 100 | Helped |
| use CAIL instead of 2IWIL on success rate | CAIL · Simulated robot arm | success rate (%): 100 -> 100 | No clear effect |
| train only on non-optimal demonstrations | CAIL · Ant | expected return | Helped |
| reduce ranking label ratio from 100% to 1% (ablation) | CAIL · Reacher | expected return: -7.8 -> -8.0 | No clear effect |
| increase ranking label ratio from 1% to 100% (ablation) | T-REX · Reacher | expected return: -83.7 -> -27.2 | Helped |
| increase ranking label ratio from 1% to 100% (ablation) | 2IWIL · Reacher | expected return: -33.5 -> -25.2 | No clear effect |
Closest papers by meaning
- Uncertainty-Based Smooth Policy Regularisation for Reinforcement Learning with Few Demonstrations (NeurIPS 2025 · similarity 0.90)
- Learning from Demonstrations via Capability-Aware Goal Sampling (NeurIPS 2025 · similarity 0.89)
- Limited Preference Aided Imitation Learning from Imperfect Demonstrations (ICML 2024 · similarity 0.89)
- Imitation Learning from Purified Demonstrations (ICML 2024 · similarity 0.89)
- GO-DICE: Goal-Conditioned Option-Aware Offline Imitation Learning via Stationary Distribution Correction Estimation (AAAI 2024 · similarity 0.88)
- Causal Imitation for Markov Decision Processes: a Partial Identification Approach (NeurIPS 2024 · 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.