Research map / Optimization / Optimizers
Differentiable Quality Diversity
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 CMA-MEGA instead of derivative-free MAP-Elites (line) | Linear Projection · LP (sphere) | QD-score: 12.21 -> 75.29 | Helped |
| use CMA-MEGA instead of derivative-free MAP-Elites (line) | Linear Projection · LP (Rastrigin) | QD-score: 8.12 -> 62.54 | Helped |
| use CMA-MEGA instead of OG-MAP-Elites (line) | Arm Repertoire · Arm Repertoire | QD-score: 59.66 -> 74.18 | Helped |
| use Adam for CMA-MEGA updates | CMA-MEGA · LSI | QD-score: 5.36 -> 21.82 | Helped |
| use Adam for CMA-MEGA updates | CMA-MEGA · LP (sphere) | QD-score: 75.29 -> 75.30 | No clear effect |
| use Adam for CMA-MEGA updates | CMA-MEGA · Arm Repertoire | QD-score: 74.18 -> 73.82 | No clear effect |
| add gradient normalization (ablation) | OMG-MEGA · LP (sphere) | QD-score: 56.96 ± 0.24 -> 71.58 ± 0.10 | Helped |
| add gradient normalization (ablation) | OMG-MEGA · LP (Rastrigin) | QD-score: 16.56 ± 0.11 -> 55.90 ± 0.21 | Helped |
| add gradient normalization (ablation) | OMG-MEGA · Arm Repertoire | QD-score: 0.85 ± 0.04 -> 44.12 ± 0.06 | Helped |
| adapt gradient coefficients with CMA-ES rather than using fixed-distribution coefficients | MEGA · LP (sphere) | QD-score: 71.58 -> 75.29 | Helped |
| adapt gradient coefficients with CMA-ES rather than using fixed-distribution coefficients | MEGA · LP (Rastrigin) | QD-score: 55.90 -> 62.54 | Helped |
Closest papers by meaning
- Quality-Diversity with Limited Resources (ICML 2024 · similarity 0.87)
- Proximal Policy Gradient Arborescence for Quality Diversity Reinforcement Learning (ICLR 2024 · similarity 0.86)
- Quality Diversity through Human Feedback: Towards Open-Ended Diversity-Driven Optimization (ICML 2024 · similarity 0.85)
- Diversity By Design: Leveraging Distribution Matching for Offline Model-Based Optimization (ICML 2025 · similarity 0.84)
- Projection-Free Variance Reduction Methods for Stochastic Constrained Multi-Level Compositional Optimization (ICML 2024 · similarity 0.84)
- Differentiation Through Black-Box Quadratic Programming Solvers (NeurIPS 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.