Genetic algorithms with DNN-based trainable crossover as an example of partial specialization of general search

TL;DR

This paper explores partial specialization of genetic algorithms by integrating a trainable neural network-based crossover operator, demonstrating improved efficiency over general GAs and discriminative models in specific tasks.

Contribution

It introduces a novel approach of using a neural network-based crossover in GAs, bridging the gap between general search and complete specialization.

Findings

Trainable crossover operators can outperform general GAs.

Partially specialized GAs can be more efficient than fully specialized neural networks.

Neural network-based crossover offers a feasible way to specialize search procedures.

Abstract

Universal induction relies on some general search procedure that is doomed to be inefficient. One possibility to achieve both generality and efficiency is to specialize this procedure w.r.t. any given narrow task. However, complete specialization that implies direct mapping from the task parameters to solutions (discriminative models) without search is not always possible. In this paper, partial specialization of general search is considered in the form of genetic algorithms (GAs) with a specialized crossover operator. We perform a feasibility study of this idea implementing such an operator in the form of a deep feedforward neural network. GAs with trainable crossover operators are compared with the result of complete specialization, which is also represented as a deep neural network. Experimental results show that specialized GAs can be more efficient than both general GAs and discriminative models.