Research map / Reinforcement learning / Visual and representation RL
Emergent Complexity and Zero-shot Transfer via Unsupervised Environment Design
Michael Dennis, Natasha Jaques, Eugene Vinitsky, Alexandre Bayen, Stuart J. Russell, Andrew Critch, 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 PAIRED (regret with antagonist) instead of minimax adversary | navigation agent · Labyrinth zero-shot transfer | success rate (%): 10 -> 40 | Improved |
| Use PAIRED instead of minimax adversary | navigation agent · Maze zero-shot transfer | success rate (%): 0.0 -> 18 | Improved |
| Use PAIRED instead of domain randomization | navigation agent · Labyrinth zero-shot transfer | success rate (%): 0.0 -> 40 | Improved |
| Use PAIRED instead of minimax or DR on easy transfer tasks (Empty, 50 Blocks) | navigation agent · Empty and 50 Blocks transfer | success rate | No clear effect |
| Use PAIRED instead of minimax adversary after adversary constraints are removed | MuJoCo hopper agent · MuJoCo hopper with adversarial torques | reward | Improved |
| Compute regret with combined population or flexible approach that breaks the coordination assumption (ablation) | PAIRED · Navigation emergent complexity | solved path length | Hurt |
Closest papers by meaning
- Adversarial Environment Design via Regret-Guided Diffusion Models (NeurIPS 2024 · similarity 0.92)
- Improving Regret Approximation for Unsupervised Dynamic Environment Generation (NeurIPS 2025 · similarity 0.91)
- Marginal Benefit Driven RL Teacher for Unsupervised Environment Design (AAAI 2025 · similarity 0.90)
- Refining Minimax Regret for Unsupervised Environment Design (ICML 2024 · similarity 0.89)
- No Regrets: Investigating and Improving Regret Approximations for Curriculum Discovery (NeurIPS 2024 · similarity 0.89)
- DRED: Zero-Shot Transfer in Reinforcement Learning via Data-Regularised Environment Design (ICML 2024 · similarity 0.88)
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.
