Evolutionary Computation Meets Machine Learning: A Survey

  title={Evolutionary Computation Meets Machine Learning: A Survey},
  author={Jun Zhang and Zhi-hui Zhan and Ying Lin and Ni Chen and Yue-jiao Gong and Jinghui Zhong and Henry Shu-hung Chung and Yun Li and Yu-hui Shi},
  journal={IEEE Computational Intelligence Magazine},
Evolutionary computation (EC) is a kind of optimization methodology inspired by the mechanisms of biological evolution and behaviors of living organisms. In the literature, the terminology evolutionary algorithms is frequently treated the same as EC. This article focuses on making a survey of researches based on using ML techniques to enhance EC algorithms. In the framework of an ML-technique enhanced-EC algorithm (MLEC), the main idea is that the EC algorithm has stored ample data about the… 

Figures from this paper

A survey on algorithm adaptation in evolutionary computation

This paper presents a classification of adaptive EC (AEC) algorithms from the perspective of how an adaptation scheme is designed, involving the adaptation objects, adaptation evidences, and adaptation methods, by analyzing the population distribution characteristics of EC algorithms.

Enhancing the performance of evolutionary algorithms: A novel maturity-based adaptation strategy

Experimental results on eight benchmark functions show that the proposed maturity-based adaptation strategy for EAs can bring significant improvements in search speed, solution accuracy and robustness.

An Energy-Based Sampling Technique for Multi-Objective Restricted Boltzmann Machine

This paper examines the sampling techniques of a restricted Boltzmann machine-based multi-objective (MO) estimation of distribution algorithm (REDA) and proposes a sampling mechanism that exploits the energy information of the solutions in a trained network to improve the search capability of the algorithm.

Analysis of Available Selection Techniques and Recommendation for Memetic Algorithm and its Application to TSP

This research mainly focuses on studying and evaluating the selection techniques and steady state selection technique shows better results than the existing selection operator.

A Hybrid Backtracking Search Optimization Algorithm with Differential Evolution

A hybrid backtracking search optimization algorithm with differential evolution, called HBD, which is used to accelerate the convergence by optimizing one worse individual according to its probability at each iteration process.

A survey on evolutionary machine learning

This paper provides a review on evolutionary machine learning techniques for major machine learning tasks such as classification, regression and clustering, and emerging topics including combinatorial optimisation, computer vision, deep learning, transfer learning, and ensemble learning.

Adaptive genetic algorithm based on density distribution of population

A parameter adaptation strategy which aims at evaluating the density distribution of population as well as both the fitness values comprehensively, and adapting the parameters accordingly is proposed.

Application of on-line machine learning in optimization algorithms: A case study for local search

  • Cong HaoT. Yoshimura
  • Computer Science
    2017 9th Computer Science and Electronic Engineering (CEEC)
  • 2017
An on-line learning based estimator to perform the solution estimation in heuristic algorithms to improve estimation accuracy and the accuracy of this estimator is much higher than the static estimator, and is also higher than a general off-line pre-trained learner.



A Method for Integrating Genetic Search Based Function Optimization and Competitive Learning

This paper presents an empirical analysis of where the proposed technique will outperform genetic algorithms, and describes a class of problems in which a genetic algorithm may be able to perform better.

Accelerating Differential Evolution Using an Adaptive Local Search

  • N. NomanH. Iba
  • Computer Science
    IEEE Transactions on Evolutionary Computation
  • 2008
It is shown that the proposed new version of DE, with the adaptive LS, performs better, or at least comparably, to classic DE algorithm.

Reducing Fitness Evaluations Using Clustering Techniques and Neural Network Ensembles

The k-means method is applied to group the individuals of a population into a number of clusters and shows better performance than the strategy where only the best individuals according to the approximate model are re-evaluated.

Hybridization of evolutionary algorithms and local search by means of a clustering method

This paper proposes the combination of an EA, a clustering process, and a local-search procedure to the evolutionary design of product-units neural networks and shows a favorable performance when the regression method proposed is compared to other standard methods.

Hybrid Evolutionary Search Method Based on Clusters

  • Ming LiH. Tam
  • Computer Science
    IEEE Trans. Pattern Anal. Mach. Intell.
  • 2001
Presents a hybrid evolutionary search method based on clusters (HESC). The method is specifically designed to enhance the search efficiency while alleviating the problem of premature convergence

Clustering-Based Adaptive Crossover and Mutation Probabilities for Genetic Algorithms

The use of fuzzy logic to adaptively adjust the values of px and pm in GA is presented and the effectiveness of the fuzzy-controlled crossover and mutation probabilities is demonstrated by optimizing eight multidimensional mathematical functions.

Opposition-Based Differential Evolution

This paper presents a novel algorithm to accelerate the differential evolution (DE), which employs opposition-based learning (OBL) for population initialization and also for generation jumping and results confirm that the ODE outperforms the original DE and FADE in terms of convergence speed and solution accuracy.

A Probabilistic Memetic Framework

A novel probabilistic memetic framework is presented that models MAs as a process involving the decision of embracing the separate actions of evolution or individual learning and analyzing the probability of each process in locating the global optimum.

JADE: Adaptive Differential Evolution With Optional External Archive

Simulation results show that JADE is better than, or at least comparable to, other classic or adaptive DE algorithms, the canonical particle swarm optimization, and other evolutionary algorithms from the literature in terms of convergence performance for a set of 20 benchmark problems.

Differential Evolution Algorithm With Strategy Adaptation for Global Numerical Optimization

This paper proposes a self- Adaptive DE (SaDE) algorithm, in which both trial vector generation strategies and their associated control parameter values are gradually self-adapted by learning from their previous experiences in generating promising solutions.