Genetic Algorithms for Evolving Deep Neural Networks

TL;DR

This paper introduces a genetic algorithm-assisted method to enhance deep autoencoders, resulting in sparser neural networks and improved performance, building on prior genetic algorithm applications in neural network training.

Contribution

It presents a novel GA-assisted approach specifically designed for deep learning, extending previous genetic algorithm methods for neural network training.

Findings

GA-assisted method improves deep autoencoder performance

Produces sparser neural networks

Enhances deep learning models' efficiency

Abstract

In recent years, deep learning methods applying unsupervised learning to train deep layers of neural networks have achieved remarkable results in numerous fields. In the past, many genetic algorithms based methods have been successfully applied to training neural networks. In this paper, we extend previous work and propose a GA-assisted method for deep learning. Our experimental results indicate that this GA-assisted approach improves the performance of a deep autoencoder, producing a sparser neural network.