News and events
12 October 2011
Markov Networks and Fitness Modelling in Evolutionary Algorithms
Presented By Dr Sandy Brownlee, Research Associate, Department of Civil & Building, Loughborough University
- Room A234, Schofield Building
About this event
Abstract; A well-known paradigm for optimation is the genetic or evolutionary algorithm (EA). An EA maintains a population of possible solutions to a problem, and converges on a global optimum using biologically-inspired selection and reproduction operators. These algorithms have been shown to perform well on a variety of hard optimisation and search problems.
A recent development in evolutionary computation is the Estimation of Distribution Algorithm (EDA) which replaces the traditional genetic reproduction operators (crossover and mutation) with the construction and sampling of a probabilistic model. While this can often represent a significant computational expense, a benefit is that the model contains explicit information about the fitness function which can aid problem solving. This talk will look at one approach using a Markov network to model the distribution of fitness within the population. It will go on to explain how the resulting model can be used to reveal underlying dynamics of the problem and investigate how factors such as selection and population size affect the quality of the model.