Chapter 6 - Evolving and Spiking Connectionist Systems for Brain-Inspired Artificial Intelligence

Abstract

Artificial neural networks have now a long history as major techniques in computational intelligence with a wide range of application for learning from data and for artificial intelligence (AI). This chapter starts with a brief review of AI methods, from Aristotle's logic to the classical artificial neural networks (ANN) and hybrid systems that are used for AI now. A main focus of the paper is a class of ANN called evolving connectionist systems (ECOS) that evolve their structure and functionality through learning from data. Methods, systems, and a wide range of applications of ECOS are presented and referenced. Principles and applications of spiking neural networks (SNN) and evolving SNN (eSNN) are presented and illustrated as the third generation of ANN. The chapter includes a section on future brain-like SNN for brain-like AI. An example is the NeuCube SNN architecture, which is illustrated with deep learning algorithms and with two case study applications for brain data and seismic data modeling. The chapter concludes that knowing and combining various methods of AI and ANN and getting more inspiration from neuroscience to create new methods is the way for future research.