Model-based evolutionary optimization

Authors:
Yongqiang Wang;Michael C. Fu;Steven I. Marcus
Affiliations:
University of Maryland, MD;University of Maryland, MD;University of Maryland, MD
Venue:
Proceedings of the Winter Simulation Conference
Year:
2010

Citing 8
Cited 0

From Recombination of Genes to the Estimation of Distributions I. Binary Parameters

PPSN IV Proceedings of the 4th International Conference on Parallel Problem Solving from Nature
A Fictitious Play Approach to Large-Scale Optimization

Operations Research
What evolutionary game theory tells us about multiagent learning

Artificial Intelligence
A particle filtering framework for randomized optimization algorithms

Proceedings of the 40th Conference on Winter Simulation
A Decentralized Approach to Discrete Optimization via Simulation: Application to Network Flow

Operations Research
A Model Reference Adaptive Search Method for Global Optimization

Operations Research
Dynamic sample budget allocation in model-based optimization

Journal of Global Optimization
On the convergence of a class of estimation of distribution algorithms

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

We propose a new framework for global optimization by building a connection between global optimization problems and evolutionary games. Based on this connection, we propose a Model-based Evolutionary Optimization (MEO) algorithm, which uses probabilistic models to generate new candidate solutions and uses various dynamics from evolutionary game theory to govern the evolution of the probabilistic models. The MEO algorithm also gives new insight into the mechanism of model updating in model-based global optimization algorithms. Based on the MEO algorithm, a novel Population Model-based Evolutionary Optimization (PMEO) algorithm is proposed, which better captures the multimodal property of global optimization problems and gives better simulation results.