Research map / Optimization / Optimizers
EvoGrad: Evolutionary-Weighted Gradient and Hessian Learning for Black-Box Optimization
Yedidya Kfir, Elad Sarafian, Yoram Louzoun, Sarit Kraus
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 |
|---|---|---|---|
| add evolutionary (CMA covariance) weighting to gradient learning | EGL · COCO suite | solved functions: 0.821 -> 0.917 | Improved |
| add Hessian (Jacobian) correction to gradient learning | EGL · COCO suite | solved functions: 0.821 -> 0.858 | Improved |
| combine evolutionary weighting and Hessian correction | EvoGrad · COCO suite | solved functions: 0.917 -> 0.949 | Improved |
| replace CMA covariance weights with softmax weights (ablation) | EvoGrad · COCO suite | solved functions: 0.917 -> 0.71 | Hurt |
| attach Jacobian to computation graph (full backprop) (ablation) | HGrad · COCO suite | solved functions: 0.858 vs 0.858 | No clear effect |
| add trust region to CMA | CMA · COCO suite | solved functions: 0.612 -> 0.740 | Improved |
| vary network structure (ablation) | EvoGrad2 · hyperparameter sweep | coefficient of variation: CV 0.0429 | No clear effect |
| seed EvoGrad with CMA solution | EvoGrad · Adversarial attack | convergence time: 6H -> 3H | Improved |
| use EvoGrad instead of CMA | CMA · Adversarial attack | MSE: 0.05 -> 0.001 | Improved |
Closest papers by meaning
- MetaBox-v2: A Unified Benchmark Platform for Meta-Black-Box Optimization (NeurIPS 2025 · similarity 0.87)
- Posterior Inference in Latent Space for Scalable Constrained Black-box Optimization (NeurIPS 2026 · similarity 0.86)
- Posterior Inference with Diffusion Models for High-dimensional Black-box Optimization (ICML 2025 · similarity 0.86)
- LPGD: A General Framework for Backpropagation through Embedded Optimization Layers (ICML 2024 · similarity 0.86)
- EVOS: Efficient Implicit Neural Training via EVOlutionary Selector (CVPR 2025 · similarity 0.85)
- Learning Surrogates for Offline Black-Box Optimization via Gradient Matching (ICML 2024 · similarity 0.85)
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.
