Research map / Reinforcement learning / Imitation learning
Cross-domain Imitation from Observations
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 inference task adaptation (ablation) | xDIO · M-A2A | normalized reward: 0.46\pm0.12 -> 0.78\pm0.08 | Helped |
| add inference task adaptation (ablation) | xDIO · M-A2C | normalized reward: 0.78\pm0.23 -> 1.00\pm0.00 | Helped |
| add the mutual information objective (ablation) | xDIO · M-A2A | normalized reward: 0.28\pm0.20 -> 0.78\pm0.08 | Helped |
| add the mutual information objective (ablation) | xDIO · M-A2C | normalized reward: 0.80\pm0.21 -> 1.00\pm0.00 | Helped |
| add temporal position preservation (ablation) | xDIO · M-A2A | normalized reward: 0.36\pm0.50 -> 0.78\pm0.08 | Helped |
| add temporal position preservation (ablation) | xDIO · M-A2C | normalized reward: 0.43\pm0.50 -> 1.00\pm0.00 | Helped |
| use xDIO rather than IF | V-R2R cross-domain imitation · V-R2R | normalized reward: 0.32\pm0.10 -> 0.95\pm0.03 | Helped |
| use xDIO rather than CCA | V-R2W cross-domain imitation · V-R2W | normalized reward: 0.86\pm0.30 -> 0.93\pm0.01 | Helped |
| use xDIO rather than IF | D-R2R cross-domain imitation · D-R2R | normalized reward: 0.48\pm0.30 -> 0.99\pm0.02 | Helped |
| use xDIO rather than IF | M-R2R cross-domain imitation · M-R2R | normalized reward: 0.61\pm0.23 -> 0.96\pm0.07 | Helped |
| use xDIO rather than CCA | M-A2A cross-domain imitation · M-A2A | normalized reward: 0.30\pm0.30 -> 0.78\pm0.08 | Helped |
| use xDIO rather than CycleGAN | M-A2C cross-domain imitation · M-A2C | normalized reward: 0.80\pm0.28 -> 1.00\pm0.00 | Helped |
Closest papers by meaning
- Learn what matters: cross-domain imitation learning with task-relevant embeddings (NeurIPS 2022 · similarity 0.92)
- Causal Imitation for Markov Decision Processes: a Partial Identification Approach (NeurIPS 2024 · similarity 0.92)
- Domain Adaptive Imitation Learning with Visual Observation (NeurIPS 2023 · similarity 0.91)
- Agnostic Interactive Imitation Learning: New Theory and Practical Algorithms (ICML 2024 · similarity 0.89)
- Interactive and Hybrid Imitation Learning: Provably Beating Behavior Cloning (NeurIPS 2025 · similarity 0.89)
- Imitation Learning from Observation with Automatic Discount Scheduling (ICLR 2024 · 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.