Research map / Reinforcement learning / MDPs and planning
AutoQD: Automatic Discovery of Diverse Behaviors with Quality-Diversity Optimization
Saeed Hedayatian, Stefanos Nikolaidis
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 learned random-feature policy embeddings with weighted PCA descriptors instead of hand-crafted descriptors | CMA-MAE QD · Ant | GT QD Score (x10^4): 182.58 -> 361.43 | Improved |
| use AutoQD instead of hand-crafted descriptors | CMA-MAE QD · BipedalWalker | GT QD Score (x10^4): 1.81 -> 6.09 | Improved |
| use AutoQD instead of hand-crafted descriptors | CMA-MAE QD · Swimmer | Vendi Score: 4.67 -> 16.92 | Improved |
| use AutoQD instead of hand-crafted descriptors | CMA-MAE QD · Walker2d | qVS: 9.08 -> 7.22 | Hurt |
| use AutoQD instead of hand-crafted descriptors | CMA-MAE QD · Walker2d | Vendi Score: 10.17 -> 8.4 | Hurt |
| use AutoQD instead of hand-crafted descriptors | CMA-MAE QD · HalfCheetah | qVS: 2.07 -> 1.35 | Hurt |
| use AutoQD instead of Aurora learned descriptors | CMA-MAE QD · HalfCheetah | Vendi Score: 5.8 -> 5.29 | Hurt |
| use AutoQD instead of hand-crafted descriptors | CMA-MAE QD · HalfCheetah | GT QD Score (x10^4): 24.91 -> 30.78 | Improved |
| use AutoQD instead of hand-crafted descriptors | CMA-MAE QD · Hopper | qVS: 1.35 -> 1.94 | Improved |
| use AutoQD population under altered friction | best policy · BipedalWalker (varying friction) | AUC: 1309.41 -> 1429.66 | Improved |
| use AutoQD population under altered mass | best policy · BipedalWalker (varying mass scale) | AUC: 271.83 -> 295.65 | Improved |
| use AutoQD instead of RegularQD under altered friction | best policy · BipedalWalker (varying friction) | AUC: 30.27 -> 1429.66 | Improved |
Closest papers by meaning
- Behavioral Foundation Models for Quality Diversity (NeurIPS 2026 · similarity 0.90)
- Proximal Policy Gradient Arborescence for Quality Diversity Reinforcement Learning (ICLR 2024 · similarity 0.87)
- Quality Diversity through Human Feedback: Towards Open-Ended Diversity-Driven Optimization (ICML 2024 · similarity 0.87)
- Offline Behavior Distillation (NeurIPS 2024 · similarity 0.87)
- Iteratively Refined Behavior Regularization for Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.86)
- Quality-Diversity Actor-Critic: Learning High-Performing and Diverse Behaviors via Value and Successor Features Critics (ICML 2024 · similarity 0.86)
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.
