Review of Advances in Neural Networks: Neural Design Technology Stack

Abstract

This review provides a high-level synthesis of significant recent advances in artificial neural network research. We assume that a global view of the field can benefit researchers by providing alternative viewpoints. Therefore, we present different network and neuron models, we discuss model parameters and the means to obtain them, and we draw a quick outline of information encoding, before proceeding to an overview of the relevant learning mechanisms, ranging from established approaches to novel ideas. We specifically focus on comparing the classical artificial model with the biologically-feasible spiking neuron.