The dynamics of a genetic algorithm for a simple learning problem

DOI10.1088/0305-4470/29/23/013MaRDI QIDQ4393793zbMATH OpenOpenAlexWikidataFDO

Authors Magnus Rattray, Jonathan L. Shapiro

Publication date 2 March 1999

Published in Journal of Physics A: Mathematical and General (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/cond-mat/9609109, https://publications.aston.ac.uk/id/eprint/686/1/misc96-010.pdf

zbMATH Keywords

genetic algorithms

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Neural nets applied to problems in time-dependent statistical mechanics (82C32)

Abstract: A formalism for describing the dynamics of Genetic Algorithms (GAs) using methods from statistical mechanics is applied to the problem of generalization in a perceptron with binary weights. The dynamics are solved for the case where a new batch of training patterns is presented to each population member each generation, which considerably simplifies the calculation. The theory is shown to agree closely to simulations of a real GA averaged over many runs, accurately predicting the mean best solution found. For weak selection and large problem size the difference equations describing the dynamics can be expressed analytically and we find that the effects of noise due to the finite size of each training batch can be removed by increasing the population size appropriately. If this population resizing is used, one can deduce the most computationally efficient size of training batch each generation. For independent patterns this choice also gives the minimum total number of training patterns used. Although using independent patterns is a very inefficient use of training patterns in general, this work may also prove useful for determining the optimum batch size in the case where patterns are recycled.

Recommendations

• scientific article; zbMATH DE number 1664949
• scientific article; zbMATH DE number 1462668
• Dynamical modelling of a genetic algorithm
• A simple model for genetic algorithm convergence
• scientific article; zbMATH DE number 1381974
• Learning and complexity in genetic auto-adaptive systems
• scientific article; zbMATH DE number 1873257

Cited in

(9)

• Adaptive learning by genetic algorithms. Analytic results and applications to economic models
• The dynamics of a genetic algorithm for simple random Ising systems
• Cumulant dynamics of a population under multiplicative selection, mutation, and drift.
• scientific article; zbMATH DE number 6177874 (Why is no real title available?)
• Finite populations induce metastability in evolutionary search.
• Learning dynamics, genetic algorithms, and corporate takeovers
• Scaling, machine learning, and genetic neural nets
• When a genetic algorithm outperforms hill-climbing
• Optimizing epochal evolutionary search: population-size independent theory.

This page was built for publication: The dynamics of a genetic algorithm for a simple learning problem

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4393793)