Research map / Learning theory and algorithms / Matrix and tensor methods
CMA-ES with Optimal Covariance Update and Storage Complexity
Oswin Krause, Dídac Rodríguez Arbonès, Christian Igel
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 |
|---|---|---|---|
| Replace eigendecomposition covariance update with Cholesky factor update | CMA-ES · Benchmark functions, d=64 | wall-clock time: 20 times faster | Improved |
| Use Cholesky-CMA-ES instead of CMA-ES-Ref | CMA-ES · Benchmark functions | function evaluations | No clear effect |
| Delay covariance update by d steps (CMA-ES/d) instead of per-iteration update | CMA-ES-Ref · Benchmark functions | function evaluations | Hurt |
| Increase dimensionality | Cholesky-CMA-ES speed advantage · Benchmark functions, d 4 to 256 | wall-clock speed-up | Improved |
| Use Cholesky-CMA-ES instead of CMA-ES/d and Suttorp-CMA-ES | CMA-ES · Benchmark functions | wall-clock time | Improved |
| Grow dimension to d = 256 so matrix exceeds L2 cache | Cholesky-CMA-ES and Suttorp-CMA-ES · Benchmark functions, d=256 | runtime | Hurt |
Closest papers by meaning
- Towards Understanding the Dynamics of Low-Rank Adaptation (ICML 2026 · similarity 0.84)
- Efficient Globally Convergent Stochastic Optimization for Canonical Correlation Analysis (NeurIPS 2016 · similarity 0.84)
- Compress to Impress: Efficient LLM Adaptation Using a Single Gradient Step on 100 Samples (NeurIPS 2025 · similarity 0.83)
- Bayesian Weight Enhancement with Steady-State Adaptation for Test-time Adaptation in Dynamic Environments (ICML 2025 · similarity 0.83)
- RCCDA: Adaptive Model Updates in the Presence of Concept Drift under a Constrained Resource Budget (NeurIPS 2025 · similarity 0.82)
- Unconstrained Stochastic CCA: Unifying Multiview and Self-Supervised Learning (ICLR 2024 · similarity 0.82)
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.
