Research map / Reinforcement learning / Imitation learning
Invariant Causal Imitation Learning for Generalizable Policies
Ioana Bica, Daniel Jarrett, Mihaela van der Schaar
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 |
|---|---|---|---|
| ICIL instead of BC | Imitation policy · MIMIC-III mechanical ventilator | ACC: 0.783 ± 0.001 -> 0.855 ± 0.003 | Improved |
| ICIL instead of VDICE, the best baseline | Imitation policy · MIMIC-III mechanical ventilator | AUC: 0.784 ± 0.001 -> 0.856 ± 0.004 | Improved |
| ICIL instead of VDICE, the best baseline | Imitation policy · MIMIC-III mechanical ventilator | APR: 0.716 ± 0.001 -> 0.789 ± 0.004 | Improved |
| ICIL instead of EDM | Imitation policy · MIMIC-III mechanical ventilator | AUC: 0.741 ± 0.011 -> 0.856 ± 0.004 | Improved |
| Add IRMv1 penalty to BC | BC · MIMIC-III mechanical ventilator | ACC: 0.783 ± 0.001 -> 0.791 ± 0.002 | No clear effect |
| Add IRMv1 penalty to RCAL | RCAL · MIMIC-III mechanical ventilator | AUC: 0.771 ± 0.002 -> 0.766 ± 0.003 | No clear effect |
| Add IRMv1 penalty to VDICE | VDICE · MIMIC-III mechanical ventilator | ACC: 0.794 ± 0.001 -> 0.766 ± 0.001 | Hurt |
| Add IRMv1 penalty to VDICE | VDICE · MIMIC-III mechanical ventilator | AUC: 0.784 ± 0.001 -> 0.730 ± 0.001 | Hurt |
| Add IRMv1 penalty to EDM | EDM · MIMIC-III mechanical ventilator | AUC: 0.741 ± 0.011 -> 0.717 ± 0.015 | Hurt |
| Add IRMv1 penalty to EDM | EDM · MIMIC-III mechanical ventilator | APR: 0.682 ± 0.005 -> 0.673 ± 0.007 | Hurt |
| Add IRMv1 invariance penalty to existing batch imitation methods | BC, RCAL, VDICE, EDM · OpenAI gym tasks (Acrobot, Cartpole, LunarLander, BeamRider) | average return | No clear effect |
| ICIL instead of all benchmarks | Imitation policy · OpenAI gym tasks (Acrobot, Cartpole, LunarLander, BeamRider) | average return on unseen test environment | Improved |
Closest papers by meaning
- Imitating Latent Policies from Observation (ICML 2019 · similarity 0.91)
- Causal Imitation Learning via Inverse Reinforcement Learning (ICLR 2023 · similarity 0.90)
- Causal Imitation Learning under Expert-Observable and Expert-Unobservable Confounding (ICLR 2026 · similarity 0.90)
- Causal Imitation for Markov Decision Processes: a Partial Identification Approach (NeurIPS 2024 · similarity 0.90)
- Learning When to Stop: Selective Imitation Learning Under Arbitrary Dynamics Shift (NeurIPS 2026 · similarity 0.88)
- Instant Policy: In-Context Imitation Learning via Graph Diffusion (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.
