The Boon of Gene-Culture Interaction for Effective Evolutionary Multitasking
Multifactorial optimization (MFO) is a recently proposed paradigm for evolutionary multitasking that is inspired by the possibility of harnessing underlying synergies between outwardly unrelated optimization problems through the process of implicit genetic transfer. In contrast to traditional single-objective and multi-objective optimization, which consider only a single problem in one optimization run, MFO aims at solving multiple optimization problems simultaneously. Through comprehensive empirical study, MFO has demonstrated notable performance on a variety of complex optimization problems. In this paper, we take a step towards better understanding the means by which MFO leads to the observed performance improvement. In particular, since (a) genetic and (b) cultural transmission across generations form the crux of the proposed evolutionary multitasking engine, we focus on how their interaction (i.e., gene-culture interaction) affects the overall efficacy of this novel paradigm.
KeywordsAssortative Mating Benchmark Function Search Region Optimization Task Cultural Bias
This work was conducted within the Rolls-Royce@NTU Corporate Lab with support from the National Research Foundation (NRF) Singapore under the Corp Lab@University Scheme.
- 6.Ishibuchi, H., Tsukamoto, N., Nojima, Y.: Evolutionary many-objective optimization: a short review. In: IEEE Congress on Evolutionary Computation, pp. 2419–2426. Citeseer (2008)Google Scholar
- 8.Rice, J., Cloninger, C.R., Reich, T.: Multifactorial inheritance with cultural transmission and assortative mating. I. Description and basic properties of the unitary models. Am. J. Hum. Genet. 30(6), 618 (1978)Google Scholar
- 16.Ong, Y.S., Zhou, Z., Lim, D.: Curse and blessing of uncertainty in evolutionary algorithm using approximation. In: 2006 IEEE Congress on Evolutionary Computation, CEC 2006, pp. 2928–2935. IEEE (2006)Google Scholar