An Artificial Neural Network Functionalized by Evolution

TL;DR

This paper introduces a hybrid neural network model that combines tensor calculus with evolutionary mechanisms to discover efficient topologies for complex AI tasks, enabling early adaptation and structural convergence.

Contribution

It presents a novel hybrid approach integrating tensor calculus and pseudo-Darwinian evolution to optimize neural network topologies for specific AI applications.

Findings

Able to find well-adapted topologies early in evolution

Demonstrates structural convergence in neural network design

Applicable to robotics, big data, and artificial life

Abstract

The topology of artificial neural networks has a significant effect on their performance. Characterizing efficient topology is a field of promising research in Artificial Intelligence. However, it is not a trivial task and it is mainly experimented on through convolutional neural networks. We propose a hybrid model which combines the tensor calculus of feed-forward neural networks with Pseudo-Darwinian mechanisms. This allows for finding topologies that are well adapted for elaboration of strategies, control problems or pattern recognition tasks. In particular, the model can provide adapted topologies at early evolutionary stages, and 'structural convergence', which can found applications in robotics, big-data and artificial life.