{"paper":{"title":"The CMA Evolution Strategy: A Tutorial","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"The CMA-ES derives its search distribution from intuitive requirements for non-linear non-convex optimization in continuous space.","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Nikolaus Hansen (TAO)","submitted_at":"2016-04-04T08:16:12Z","abstract_excerpt":"This tutorial introduces the CMA Evolution Strategy (ES), where CMA stands for Covariance Matrix Adaptation. The CMA-ES is a stochastic, or randomized, method for real-parameter (continuous domain) optimization of non-linear, non-convex functions. We try to motivate and derive the algorithm from intuitive concepts and from requirements of non-linear, non-convex search in continuous domain."},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The CMA-ES is a stochastic method for real-parameter optimization of non-linear, non-convex functions, motivated and derived from intuitive concepts and requirements of non-linear non-convex search in continuous domain.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The reader possesses sufficient background in basic probability, linear algebra, and optimization to follow the intuitive derivation of the covariance adaptation rules.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"The CMA-ES is a stochastic optimizer for continuous non-linear non-convex problems that adapts a multivariate normal search distribution using covariance matrix updates derived from intuitive requirements.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"The CMA-ES derives its search distribution from intuitive requirements for non-linear non-convex optimization in continuous space.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"10dde5b5d707cc142ac3734513459c8b4cc4d1a8153e4a58cb3ec42fc4cc270f"},"source":{"id":"1604.00772","kind":"arxiv","version":2},"verdict":{"id":"e468d608-345b-44aa-98b0-7d7a1d6be5df","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T21:06:33.599420Z","strongest_claim":"The CMA-ES is a stochastic method for real-parameter optimization of non-linear, non-convex functions, motivated and derived from intuitive concepts and requirements of non-linear non-convex search in continuous domain.","one_line_summary":"The CMA-ES is a stochastic optimizer for continuous non-linear non-convex problems that adapts a multivariate normal search distribution using covariance matrix updates derived from intuitive requirements.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The reader possesses sufficient background in basic probability, linear algebra, and optimization to follow the intuitive derivation of the covariance adaptation rules.","pith_extraction_headline":"The CMA-ES derives its search distribution from intuitive requirements for non-linear non-convex optimization in continuous space."},"references":{"count":37,"sample":[{"doi":"","year":2017,"title":"Quality gain analysis of the weighted recombination evolution strategy on general convex quadratic functions","work_id":"5504e6eb-107f-4dce-b8d4-d7de66e41a05","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Diagonal acceleration for covariance matrix adaptation evolution strategies","work_id":"01b4c5ef-72d2-43b7-ae97-226946381ae4","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2005,"title":"A restart CMA evolution strategy with increasing population size","work_id":"bd06d30d-492a-4b76-acb9-6121cfaffe3c","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2006,"title":"Weighted multirecombination evolution strategies","work_id":"d389d83a-1029-45ab-b0d4-7f09748f0ec2","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2004,"title":"Performance analysis of evolutionary optimization with cumu- lative step length adaptation","work_id":"812a94cd-a8b9-4f58-a375-34e27afc2874","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":37,"snapshot_sha256":"82f86b7d9949e5301b1de07b013555cded0741126b9e97c9cc29b627b031aa01","internal_anchors":1},"formal_canon":{"evidence_count":3,"snapshot_sha256":"ef629d0c198d8421d6496899e84fc961d8d4d6d9c51f5885483ecce24d3354c4"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}