Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Continual Learning of Control Primitives : Skill Discovery via Reset-Games
Kelvin Xu, Siddharth Verma, Chelsea Finn, Sergey Levine
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 |
|---|---|---|---|
| use multiple resetter skills (reset game with diversity bonus) instead of a single adversary (ablation) | LSR · DClaw-ManipulateFreeObject | reset-free performance | Improved |
| remove the diversity bonus (Multiple-Resetters variant) (ablation) | LSR · DClaw-ManipulateFreeObject | reset-free performance | Hurt |
| use LSR instead of R3L perturbation controller | reset-free learning · DClaw-ManipulateFreeObject | iterations to converge: ~28.6% faster | Improved |
| use an explicit reset policy instead of reset skill game | reset-free learning · DClaw-ManipulateFreeObject | performance | Hurt |
| use LSR skills instead of DIAYN and DADS skills | hierarchical controller · Ant-MediumMaze | normalized return | Improved |
| use LSR skills instead of DIAYN and DADS skills | hierarchical controller · Ant-Waypoints | normalized return | Improved |
| use (x,y) prior state representation | DIAYN and DADS · Ant skill learning | skill quality | Mixed |
Closest papers by meaning
- Continual Reinforcement Learning by Planning with Online World Models (ICML 2025 · similarity 0.85)
- When Do Skills Help Reinforcement Learning? A Theoretical Analysis of Temporal Abstractions (ICML 2024 · similarity 0.85)
- Environment Design for Inverse Reinforcement Learning (ICML 2024 · similarity 0.84)
- The Power of Resets in Online Reinforcement Learning (NeurIPS 2024 · similarity 0.83)
- DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning (NeurIPS 2025 · similarity 0.83)
- A Single Goal is All You Need: Skills and Exploration Emerge from Contrastive RL without Rewards, Demonstrations, or Subgoals (ICLR 2025 · similarity 0.83)
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.
