cmaartic.pdf

N/A

Completely Derandomized Self-Adaptation in Evolution Strategies Nikolaus Hansen and Andreas Ostermeier In Evolutionary Computation, 9(2), pp. 159-195 (2001) Errata • Section 3, footnote 9: ”We use the expectation of nk ln(...)” must be ”We use the expectation of 21 nk ln(...)”. • (same error) Section 3, same page, description Performance: ”...where ln(...)...” must be ”...where 21 ln(...)...”. Additional Comments • Section 5.1, Table 1: The default parameter setting for wi makes sense only, if µ ≤ λ/2. Otherwise weights…

related reading

• A Visual Guide to Evolution Strategies | 大トロblog.otoro.net
• Evolution as Backstop for Reinforcement Learning · Gwern.netgwern.net
• An Introduction to Genetic Algorithms and Evolution Strate…static.aminer.org
• Illuminating the Three Dogmas of RL under Evolutionary Light (Mani Hamidi) - Sensorimotor AI Journal Clubsensorimotorai.github.io
• [2511.16652] Evolution Strategies at the Hyperscalearxiv.org
• The Practitioner's Guide to the Maximal Update Parameterization | EleutherAI Blogblog.eleuther.ai
• Competing with sampling — Alignment Research Centeralignment.org
• The Practitioner’s Guide to the Maximal Update Parameterization - Cerebrascerebras.ai
• A Mike's-Eye View of ARC's Research — Alignment Research Centeralignment.org
• Does Muon improve regulatory DNA learning? Part 1. — Origin Bioorigin.bio
• Digital Red Queen: Adversarial Program Evolution in Core War with LLMspub.sakana.ai
• Hyperparameter optimization - Wikipediaen.wikipedia.org