Research map / Reinforcement learning / Visual and representation RL
Dexterous Skill Discovery via Topology-Aware Wasserstein Dependency
James Heald, Kai Biegun, Alexandre Galashov, Maneesh Sahani
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a topology-aware Wasserstein dependency approach for unsupervised RL to discover dexterous manipulation skills with multi-fingered hands.
Problem
Unsupervised RL for dexterous object manipulation with high-dimensional multi-fingered hands; existing metric-aware skill discovery methods have not been successfully applied to such manipulators.
Approach
Argues that metric-aware skill discovery fails on high-dimensional manipulators because of metric/topology mismatch, and proposes a topology-aware Wasserstein dependency objective for skill discovery; details beyond the abstract are not available.
Closest papers by meaning
- Learning Object-Centric Motion Priors from Human for Robotic Dexterous Manipulation (AAAI 2026 · similarity 0.90)
- Scaffolding Dexterous Manipulation with Vision-Language Models (NeurIPS 2025 · similarity 0.89)
- ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning (CVPR 2025 · similarity 0.89)
- $\textit{HiMaCon:}$ Discovering Hierarchical Manipulation Concepts from Unlabeled Multi-Modal Data (NeurIPS 2025 · similarity 0.87)
- HumanoidGen: Data Generation for Bimanual Dexterous Manipulation via LLM Reasoning (NeurIPS 2025 · similarity 0.87)
- Cross-Embodiment Dexterous Grasping with Reinforcement Learning (ICLR 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.
