{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/gray-box-optimization-and-factorized","title":"Gray-box optimization and factorized distribution algorithms: where two worlds collide","arxiv_id":"1707.03093","date":"2017-07-11","proceeding":null,"authors":["Roberto Santana"],"abstract":"The concept of gray-box optimization, in juxtaposition to black-box\noptimization, revolves about the idea of exploiting the problem structure to\nimplement more efficient evolutionary algorithms (EAs). Work on factorized\ndistribution algorithms (FDAs), whose factorizations are directly derived from\nthe problem structure, has also contributed to show how exploiting the problem\nstructure produces important gains in the efficiency of EAs. In this paper we\nanalyze the general question of using problem structure in EAs focusing on\nconfronting work done in gray-box optimization with related research\naccomplished in FDAs. This contrasted analysis helps us to identify, in current\nstudies on the use problem structure in EAs, two distinct analytical\ncharacterizations of how these algorithms work. Moreover, we claim that these\ntwo characterizations collide and compete at the time of providing a coherent\nframework to investigate this type of algorithms. To illustrate this claim, we\npresent a contrasted analysis of formalisms, questions, and results produced in\nFDAs and gray-box optimization. Common underlying principles in the two\napproaches, which are usually overlooked, are identified and discussed.\nBesides, an extensive review of previous research related to different uses of\nthe problem structure in EAs is presented. The paper also elaborates on some of\nthe questions that arise when extending the use of problem structure in EAs,\nsuch as the question of evolvability, high cardinality of the variables and\nlarge definition sets, constrained and multi-objective problems, etc. Finally,\nemergent approaches that exploit neural models to capture the problem structure\nare covered.","url_abs":"http://arxiv.org/abs/1707.03093v1","url_pdf":"http://arxiv.org/pdf/1707.03093v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"gray-box-optimization-and-factorized","repo_url":"https://github.com/rsantana-isg/graybox_fda_paper","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"evolutionary-algorithms","task_name":"Evolutionary Algorithms"},{"task_slug":"two","task_name":"Vocal Bursts Valence Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}