Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learn what matters: cross-domain imitation learning with task-relevant embeddings
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 a task-relevant expert-state embedding instead of the full state | UDIL on XMagical · XMagical | reward | Helped |
| Use a task-relevant expert-state embedding instead of the full state | UDIL on cross-domain locomotion · Hopper, Walker, and HalfCheetah | distance | Helped |
| Reduce the expert embedding to the minimal task-relevant dimensions rather than increasing its size (ablation) | UDIL · Hopper and HalfCheetah | travelled distance | Helped |
| Increase the expert embedding dimension from d=3 to d=6 or all (ablation) | UDIL · Hopper and HalfCheetah | travelled distance | Hurt |
| Omit the time-invariance constraint (ablation) | UDIL · Cross-domain imitation learning | performance | Hurt |
| Train from a single expert trajectory instead of the usual set of trajectories (ablation) | UDIL · Cross-domain locomotion | travelled distance | Helped |
| Train from one trajectory rather than 20 (ablation) | UDIL, HalfCheetah imitating Hopper · HalfCheetah imitating Hopper | travelled distance | Mixed |
| Use adversarial training instead of the simplified goal-distance reward formulation (ablation) | UDIL · XMagical | reward | Mixed |
Closest papers by meaning
- Cross-domain Imitation from Observations (ICML 2021 · similarity 0.92)
- Domain-Robust Visual Imitation Learning with Mutual Information Constraints (ICLR 2021 · similarity 0.90)
- Adversarial Imitation Learning with Preferences (ICLR 2023 · similarity 0.90)
- Imitation Learning by Estimating Expertise of Demonstrators (ICML 2022 · similarity 0.90)
- Imitation Learning via Kernel Mean Embedding (AAAI 2018 · similarity 0.90)
- InfoGAIL: Interpretable Imitation Learning from Visual Demonstrations (NeurIPS 2017 · similarity 0.89)
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.