Research map / Reinforcement learning / Visual and representation RL
Learning Generalizable Skill Policy with Data-Efficient Unsupervised RL
Jongchan Park, Seungjun Oh, Seungho Baek, Yusung Kim
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 |
|---|---|---|---|
| add Complementary Information Bottleneck (CIB) (ablation) | GenDa · Quadruped-Pixels unseen Unicolor background | performance: 137 -> 268 | Improved |
| add CIB (ablation) | GenDa · Quadruped-Pixels unseen Gradation background | performance: 78 -> 206 | Improved |
| add CIB (ablation) | GenDa · Quadruped-Pixels original background | performance: 335 -> 331 | No clear effect |
| add skill relabeling (ablation) | GenDa · Quadruped-Numeric pre-training | state coverage | Improved |
| add CIB (ablation) | GenDa · Quadruped-Numeric downstream (RS, RG) | success rate | Improved |
| increase pre-training from 2M to 10M steps (ablation) | GenDa · Humanoid-Numeric MazeH | success rate: 0.14 -> 0.41 | Improved |
| increase pre-training from 2M to 10M steps (ablation) | GenDa · Humanoid-Numeric (RS, RG) | success rate: 0.77 -> 0.68 | Hurt |
| vary beta from 1.0 to 4.0 (ablation) | GenDa · Humanoid-Numeric | state coverage: 2192 -> 2595 | No clear effect |
| vary beta from 1.0 to 0.1 (ablation) | GenDa · Quadruped-Pixel | state coverage: 331 -> 286 | No clear effect |
| change c-tau mixture ratio from 0.5:0.5 to 0.25:0.75 (ablation) | GenDa relabeling · Quadruped-Numeric | state coverage: 1851 -> 1820 | No clear effect |
| raise z_roll threshold from 0.4 to 0.95 (ablation) | GenDa relabeling · Quadruped-Numeric | state coverage: 1851 -> 1678 | Hurt |
| use GenDa instead of METRA in random-start downstream | GenDa · Dog-Numeric (RS, RG) | success rate: 0.00 -> 0.53 | Improved |
Closest papers by meaning
- Foundation Policies with Hilbert Representations (ICML 2024 · similarity 0.91)
- Skill-aware Mutual Information Optimisation for Zero-shot Generalisation in Reinforcement Learning (NeurIPS 2024 · similarity 0.88)
- Robust Policy Learning via Offline Skill Diffusion (AAAI 2024 · similarity 0.88)
- Constrained Ensemble Exploration for Unsupervised Skill Discovery (ICML 2024 · similarity 0.88)
- Efficient Skill Discovery via Regret-Aware Optimization (ICML 2025 · similarity 0.87)
- Skill Disentanglement in Reproducing Kernel Hilbert Space (AAAI 2025 · 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.
