Guest Editorial Special Issue on Multitask Evolutionary Computation

It is our pleasure to introduce this special issue on multitask evolutionary computation (MTEC), focusing novel methodologies and applications of algorithms (EAs) crafted perform multiple search optimization tasks jointly. EAs are population-based methods inspired by principles natural evolution that have provided a gradient-free path solving complex learning problems. However, unlike the world where has engendered diverse species produced differently skilled subpopulations, in silico typically designed evolve set solutions specialized for just single target task. This convention problem in isolation tends curtail power xmlns:xlink="http://www.w3.org/1999/xlink">implicit parallelism population. Skills evolved given instance do not naturally transfer populations tasked solve another. Hence, convergence rates remain restrained, even settings related with overlapping spaces, similar optimal solutions, or other forms reusable information, routinely recurring.