Genetic Algorithms in Time-Dependent Environments

@article{Ronnewinkel1999GeneticAI,
  title={Genetic Algorithms in Time-Dependent Environments},
  author={Christopher Ronnewinkel and Claus O. Wilke and Thomas Martinetz},
  journal={ArXiv},
  year={1999},
  volume={physics/9911006}
}
The influence of time-dependent fitnesses on the infinite population dynamics of simple genetic algorithms (GAs) without crossover is analyzed. Based on general arguments, a schematic phase diagram is constructed that allows one to characterize the asymptotic states in dependence on the mutation rate and the time scale of changes. Furthermore, the notion of regular changes is raised for which the population can be shown to converge towards a generalized quasispecies. Based on this, error… 

Analysing fitness landscape changes in evolutionary robots

TLDR
Simulations using the exact model of the Genetic Algorithm in each problem are presented, what allows studying the dynamical behaviour of the population of the GA, helping in the analysis of the performance obtained in practice.

Digital Evolution in Time-Dependent Fitness

TLDR
The response of populations of digital organisms that adapt to a time-varying fitness landscape of two oscillating peaks is studied, corroborating in general predictions from quasi-species theory in dynamic landscapes such as adaptation to the average fitness landscape at small periods and quasistatic adaptation at large periods.

Digital Evolution in Time-Dependent Fitness Landscapes

TLDR
The response of populations of digital organisms that adapt to a time-varying fitness landscape of two oscillating peaks is studied, corroborating in general predictions from quasi-species theory in dynamic landscapes such as adaptation to the average fitness landscape at small periods and quasistatic adaptation at large periods.

Genetic Algorithms in Dynamic and Coevolving Environments

TLDR
If the GA is at all similar to natural evolution in its algorithmic flow then an investigation into how it works in dynamic and coevolving environments would prove useful for better understanding the GA.

Modeling Genetic Algorithmswith Interacting Particle SystemsP

TLDR
This work presents a natural Interacting Particle System approach for modeling and studying the asymptotic behavior of Genetic Algorithms, where a population is seen as a distribution on the search space, and the Genetic Algorithm as a measure valued dynamical system.

Modeling genetic algorithms with interacting particle systems

TLDR
A Feynman-Kac representation of the infinite population model is used to develop asymptotic stability and uniform convergence results with respect to the time parameter, which allows one to apply recent convergence results from the IPS literature for studying the convergence of genetic algorithms when the size of the population tends to infinity.

Analyzing Evolutionary Algorithms for Dynamic Optimization Problems Based on the Dynamical Systems Approach

TLDR
This chapter uses the exact model (or dynamical systems approach) to describe the standard genetic algorithm as a discrete dynamical system for DOPs and defines some properties and classes of Dops and analyzes some DOP's used by researchers in the dynamic evolutionary optimization area.

An Analysis of Dynamic Severity and Population Size

This work introduces a general mathematical framework for non-stationary fitness functions which enables the exact definition of certain problem properties. The properties' influence on the severity

References

SHOWING 1-10 OF 21 REFERENCES

Modelling the Dynamics of a Steady State Genetic Algorithm

TLDR
It is shown that, with weak selection, the steady state genetic algorithm can reproduce the dynamics of the generational genetic algorithm at half the computational cost in terms of function evaluations.

Cyclic attractors and quasispecies adaptability

TLDR
The effects of different mutation rates are examined, illustrating the ‘error threshold’ effect, that too high a mutation rate leads to a general collapse of the quasispecies and an inability to track the optimum.

Statistical Dynamics of the Royal Road Genetic Algorithm

Dynamic fitness landscapes in the quasispecies model

TLDR
The quasispecies model is studied for the special case of externally varying replication rates and the qualitative phase diagram in a given time-periodic fitness landscape can be predicted without almost any calculations.

Biological evolution through mutation, selection, and drift: An introductory review

TLDR
The terminology and recapitulate the basic models of population genetics, which describe the evolution of the composition of a population under the joint action of the various evolutionaryforces, specify the ingredients explicitly, namely, the various mutation models and fitness landscapes.

The dynamical systems model of the simple genetic algorithm

This tutorial describes the basic theory of the simple genetic algorithm, as developed by Michael Vose. The mathematical framework is established in which the actions of proportionate selection,

Generation Gaps Revisited

Evolutionary computation: comments on the history and current state

TLDR
The purpose, the general structure, and the working principles of different approaches, including genetic algorithms (GA), evolution strategies (ES), and evolutionary programming (EP) are described by analysis and comparison of their most important constituents (i.e. representations, variation operators, reproduction, and selection mechanism).

The simple genetic algorithm - foundations and theory

  • M. Vose
  • Computer Science
    Complex adaptive systems
  • 1999
TLDR
Although Michael D. Vose describes the SGA in terms of heuristic search, the book is not about search or optimization perse.

Linear Analysis of Genetic Algorithms