dorsal/arxiv
View SchemaSelf-adaptive exploration in evolutionary search
| Authors | Marc Toussaint |
|---|---|
| Categories | |
| ArXiv ID | physics/0102009 |
| URL | https://arxiv.org/abs/physics/0102009 |
Abstract
We address a primary question of computational as well as biological research on evolution: How can an exploration strategy adapt in such a way as to exploit the information gained about the problem at hand? We first introduce an integrated formalism of evolutionary search which provides a unified view on different specific approaches. On this basis we discuss the implications of indirect modeling (via a ``genotype-phenotype mapping'') on the exploration strategy. Notions such as modularity, pleiotropy and functional phenotypic complex are discussed as implications. Then, rigorously reflecting the notion of self-adaptability, we introduce a new definition that captures self-adaptability of exploration: different genotypes that map to the same phenotype may represent (also topologically) different exploration strategies; self-adaptability requires a variation of exploration strategies along such a ``neutral space''. By this definition, the concept of neutrality becomes a central concern of this paper. Finally, we present examples of these concepts: For a specific grammar-type encoding, we observe a large variability of exploration strategies for a fixed phenotype, and a self-adaptive drift towards short representations with highly structured exploration strategy that matches the ``problem's structure''.
{
"annotation_id": "28e377d3-922a-4d6b-991b-de7aa6e41298",
"date_created": "2026-03-02T18:00:32.298000Z",
"date_modified": "2026-03-02T18:00:32.298000Z",
"file_hash": "1ec5176bc42187a893ee3d5eff175fa741edf60cd049ae13e44e0c82f7819e85",
"private": false,
"record": {
"abstract": "We address a primary question of computational as well as biological research\non evolution: How can an exploration strategy adapt in such a way as to exploit\nthe information gained about the problem at hand? We first introduce an\nintegrated formalism of evolutionary search which provides a unified view on\ndifferent specific approaches. On this basis we discuss the implications of\nindirect modeling (via a ``genotype-phenotype mapping\u0027\u0027) on the exploration\nstrategy. Notions such as modularity, pleiotropy and functional phenotypic\ncomplex are discussed as implications. Then, rigorously reflecting the notion\nof self-adaptability, we introduce a new definition that captures\nself-adaptability of exploration: different genotypes that map to the same\nphenotype may represent (also topologically) different exploration strategies;\nself-adaptability requires a variation of exploration strategies along such a\n``neutral space\u0027\u0027. By this definition, the concept of neutrality becomes a\ncentral concern of this paper. Finally, we present examples of these concepts:\nFor a specific grammar-type encoding, we observe a large variability of\nexploration strategies for a fixed phenotype, and a self-adaptive drift towards\nshort representations with highly structured exploration strategy that matches\nthe ``problem\u0027s structure\u0027\u0027.",
"arxiv_id": "physics/0102009",
"authors": [
"Marc Toussaint"
],
"categories": [
"physics.bio-ph",
"cs.NE",
"nlin.AO",
"q-bio"
],
"title": "Self-adaptive exploration in evolutionary search",
"url": "https://arxiv.org/abs/physics/0102009"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "390cb00b-10f6-4cda-80bb-332ab57abcfb",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}