Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain Adaptation with Adaptive Imagination for Visual Reinforcement Learning under Limited Target Data
Hyunwoo Park, Sanghyun Lee
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 |
|---|---|---|---|
| replace CODAS with AIDA (adaptive imagination + self-consistency) | CODAS · HalfCheetah | return ratio: 0.711 -> 0.810 | Improved |
| replace CODAS with AIDA | CODAS · Hopper | return ratio: 0.356 -> 0.440 | Improved |
| replace CODAS with AIDA | CODAS · InvertedPendulum | return ratio: 0.440 -> 0.561 | Improved |
| replace CODAS with AIDA | CODAS · Walker2d | RMSE: 2.180 -> 1.784 | Improved |
| replace CODAS with AIDA | CODAS · Walker2d | return ratio: 0.038 -> 0.058 | Hurt |
| replace CODAS with AIDA | CODAS · InvertedPendulum | RMSE: 0.108 -> 0.075 | Improved |
| replace CODAS with AIDA with fewer target data | CODAS · HalfCheetah (10 target trajectories) | return ratio: 0.058 -> 0.175 | Improved |
| replace CODAS with AIDA at 30 trajectories | CODAS · HalfCheetah (30 target trajectories) | return ratio: 0.476 -> 0.573 | Improved |
| use adaptive horizon instead of fixed horizons K in {0,5,10,20} (ablation) | AIDA · Hopper | return | Improved |
| add imagination with a fixed horizon (ablation) | AIDA · Hopper | return | No clear effect |
| add self-consistency loss on imagined transitions (ablation) | AIDA · InvertedPendulum | decoded pendulum angle fidelity | Improved |
| add online-interacting PAD instead of offline AIDA | PAD · seven tasks | return ratio | Hurt |
Closest papers by meaning
- Prompt-based Visual Alignment for Zero-shot Policy Transfer (ICML 2024 · similarity 0.88)
- Fourier Guided Adaptive Adversarial Augmentation for Generalization in Visual Reinforcement Learning (AAAI 2025 · similarity 0.88)
- Revisiting Plasticity in Visual Reinforcement Learning: Data, Modules and Training Stages (ICLR 2024 · similarity 0.87)
- Solving New Tasks by Adapting Internet Video Knowledge (ICLR 2025 · similarity 0.87)
- Open-World Reinforcement Learning over Long Short-Term Imagination (ICLR 2025 · similarity 0.87)
- PRPO: Perception-Reinforced Policy Optimization via Token-Level Dynamic Advantage Reshaping (NeurIPS 2026 · similarity 0.87)
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.
