Improving adaptive differential evolution with controlled mutation strategy

Authors:
Sayan Basu Roy;Mainak Dan;Pallavi Mitra
Affiliations:
Department of Electronics and Telecommunication Engineering, Jadavpur University, Kolkata, India;Department of Electronics and Telecommunication Engineering, Jadavpur University, Kolkata, India;Department of Electronics and Telecommunication Engineering, Jadavpur University, Kolkata, India
Venue:
SEMCCO'12 Proceedings of the Third international conference on Swarm, Evolutionary, and Memetic Computing
Year:
2012

Citing 6
Cited 0

The GENITOR Algorithm and Selection Pressure: Why Rank-Based Allocation of Reproductive Trials is Best

Proceedings of the 3rd International Conference on Genetic Algorithms
A comparative study of differential evolution variants for global optimization

Proceedings of the 8th annual conference on Genetic and evolutionary computation
Differential evolution algorithm with strategy adaptation for global numerical optimization

IEEE Transactions on Evolutionary Computation
Differential evolution using a neighborhood-based mutation operator

IEEE Transactions on Evolutionary Computation
JADE: adaptive differential evolution with optional external archive

IEEE Transactions on Evolutionary Computation
Self-Adapting Control Parameters in Differential Evolution: A Comparative Study on Numerical Benchmark Problems

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we have proposed a DE variant, abbreviated by ADE_CM, to improve optimization performance of DE by imposing controlled mutation strategy. We also incorporated the concept of selective pressure to choose the random vectors in the selection of donor vector for each population. Basically we used DE/rand/1 and DE/target-to-best/1 schemes (with modifications using selective pressure) in the selection of donor using controlled mutation. The control parameter for mutation, linearly decreasing with generation, is the complement of the probability of selecting DE/target-to-best/1 in each generation. The algorithm is basically a trade-off between diversity and greediness. To improve diversity scaling factor is made adaptive and also a worst p% scheme is used in the difference vector of donor. ADE_CM is tested on 25 benchmark functions of CEC 2005 in 50 and 100 dimensions. Experimental results show that this algorithm outperforms many popular DE variants on most of the functions.