Research map / Reinforcement learning / Visual and representation RL
Reference Guided Skill Discovery
Seungeun Rho, Aaron Trinh, Danfei Xu, Sehoon Ha
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 |
|---|---|---|---|
| ground latent space with contrastive pretraining on reference motions (RGSD vs plain DIAYN) (ablation) | DIAYN · imitation on run motions | Cartesian error (cm): 52.8 -> 7.7 | Improved |
| RGSD frame-level latent reward vs GAIL-based imitation (Meta-Motivo) on fidelity | Meta-Motivo · imitation, 5 motions | Cartesian error: outperforms in 4 of 5 tasks, e.g. Backward 8.6 -> 6.7 | Improved |
| RGSD vs Meta-Motivo on naturalness | RGSD · imitation, 5 motions | Motion FID: Run 4.9 vs 9.4 (RGSD) | Hurt |
| train imitation and discovery in parallel with neighborhood latent sampling | RGSD · skill discovery near reference motions | FID stability: CALM walking 1.4 -> 15.5 vs stable RGSD | Improved |
| increase vMF concentration kappa at test time (ablation) | RGSD · three reference motions, kappa 20/100/1000 | behavioral diversity | Mixed |
| RGSD high-level policy on goal reaching | RGSD · GoalReaching freestyle | convergence speed | Improved |
| RGSD on style-conditioned goal reaching (sidestep, backward) | RGSD · GoalReaching styles | style adherence FID | Improved |
| extend RGSD with METRA backbone instead of DIAYN (ablation) | RGSD with METRA · repetitive motions | reward | Hurt |
Closest papers by meaning
- Language Guided Skill Discovery (ICLR 2025 · similarity 0.90)
- Periodic Skill Discovery (NeurIPS 2025 · similarity 0.88)
- Behavioral Exploration: Learning to Explore via In-Context Adaptation (ICML 2025 · similarity 0.87)
- GROVE: A Generalized Reward for Learning Open-Vocabulary Physical Skill (CVPR 2025 · similarity 0.87)
- METRA: Scalable Unsupervised RL with Metric-Aware Abstraction (ICLR 2024 · 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.
