Presentation 2000/1/13
Evolving Learning System Manifests Various Adaptation Mechanisms In the Dynamic Environment
Kazuko Yamasaki, Masuteru Sekiguchi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Characteristics of population under the dynamic environment for the evolving learning system and the expression of various adaptation mechanisms are studied. The Sasaki showed the differences between the Darwinian and Lamarkian type population for the first time. We examine it in simple "The Continuous Function Model"in order to investigate the mechanism of the phenomenon. Then we obtain the following results and mechanisms are briefly explained. The Darwinian type population is excellent in the adaptability, when the environment discontinuously changes, and the Lamarkian type population is excellent, when the environment continuously changes. If the same environment is continuing long, the adaptation of the Darwinian population under the dynamic environment breaks. That is the fitness may dive suddenly because of overadaptation. We also examine the population with 2 chromosomes. We obtain under various type of environments various adaptation mechanisms appear and vanish.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) evolution / learning / Dynamic Environment / multi-agent
Paper # AI99-87
Date of Issue

Conference Information
Committee AI
Conference Date 2000/1/13(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Evolving Learning System Manifests Various Adaptation Mechanisms In the Dynamic Environment
Sub Title (in English)
Keyword(1) evolution
Keyword(2) learning
Keyword(3) Dynamic Environment
Keyword(4) multi-agent
1st Author's Name Kazuko Yamasaki
1st Author's Affiliation Department of management information Tokyo university of information sciences()
2nd Author's Name Masuteru Sekiguchi
2nd Author's Affiliation Department of management information Tokyo university of information sciences
Date 2000/1/13
Paper # AI99-87
Volume (vol) vol.99
Number (no) 534
Page pp.pp.-
#Pages 8
Date of Issue