Entropy Profiles of Ranked and Random Populations

被引:0
|
作者
Milton, John [1 ]
Kennedy, Paul J. [1 ]
机构
[1] Univ Technol Sydney, Ctr Quantum Computat & Intelligent Syst, Fac Engn & IT, Sydney, NSW 2007, Australia
来源
GECCO-2010 COMPANION PUBLICATION: PROCEEDINGS OF THE 12TH ANNUAL GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE | 2010年
关键词
entropy; genetic algorithms; ranked lists; population; integer partition; information source;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The paper describes the concept of entropy profile, how it is derived, its relationship to the number partition problem and to the information extracted from an objective function. It is hoped that discussion and criticism of this idea may shed light on why some problem representations are NP hard and other, very similar problems are relatively simple. Entropy profiles illustrate the difference between ranked and randomly ordered populations of individuals in a GA in a way which quantifies the information extracted from the objective function by the ranking process. The entropy profiles of random populations are shown to arise from the fact that there are many more of such 'paths' through the entropy coordinate space than periodic or ranked paths. Additionally, entropy profile provides a measurable difference between periodic low frequency sequences, periodic high frequency sequences, random sequences and those which are in some way structured, ie by an objective function or other signal. The entropy coordinate space provides a visualisation and explanation of why these profiles differ and perhaps, by way of the integer partition phase transition, also a means to understand why some problems are hard while other seemingly similar problems are straightforward to solve.
引用
收藏
页码:1843 / 1849
页数:7
相关论文
共 50 条
  • [41] Asymptotic entropy of the ranges of random walks on discrete groups
    Xinxing Chen
    Jiansheng Xie
    Minzhi Zhao
    Science China Mathematics, 2020, 63 : 1153 - 1168
  • [42] Entropy Bounds for Discrete Random Variables via Coupling
    Sason, Igal
    2013 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY PROCEEDINGS (ISIT), 2013, : 414 - 418
  • [43] On Continuity of Range, Entropy and Drift for Random Walks on Groups
    Erschler, Anna
    RANDOM WALKS, BOUNDARIES AND SPECTRA, 2011, 64 : 55 - 64
  • [45] Asymptotic entropy of the ranges of random walks on discrete groups
    Chen, Xinxing
    Xie, Jiansheng
    Zhao, Minzhi
    SCIENCE CHINA-MATHEMATICS, 2020, 63 (06) : 1153 - 1168
  • [46] Transience of simple random walks with linear entropy growth
    Morris, Ben
    Santhakumar, Hamilton Samraj
    ELECTRONIC COMMUNICATIONS IN PROBABILITY, 2023, 28
  • [47] Learning, generalisation, and functional entropy in random automata networks
    Teuscher, Christof, 1600, Inderscience Enterprises Ltd., 29, route de Pre-Bois, Case Postale 856, CH-1215 Geneva 15, CH-1215, Switzerland (07): : 295 - 314
  • [48] EFFICIENT IMAGE CONCEPT INDEXING BY HARMONIC & ARITHMETIC PROFILES ENTROPY
    Glotin, Herve
    Zhao, Zhong-Qiu
    Ayache, Stephane
    2009 16TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOLS 1-6, 2009, : 277 - +
  • [49] Strategy to enhance entropy of random numbers in a wind-driven triboelectric random number generator
    Kim, Moon-Seok
    Tcho, Il-Woong
    Choi, Yang-Kyu
    NANO ENERGY, 2021, 89
  • [50] RANDOM WALKS AND ISOPERIMETRIC PROFILES UNDER MOMENT CONDITIONS
    Saloff-Coste, Laurent
    Zheng, Tianyi
    ANNALS OF PROBABILITY, 2016, 44 (06) : 4133 - 4183