Dendritic Computing: Branching Deeper into Machine Learning

this paper, we discuss the nonlinear computational power provided by dendrites in biological and artificial neurons. We start by briefly presenting biological evidence about the type of dendritic nonlinearities, respective plasticity rules and their effect on biological learning as assessed by computational models. Four major computational implications are identified as improved expressivity, more efficient use of resources, utilizing internal learning signals, and enabling continual learning. We then discuss examples of how dendritic computations have been used to solve real-world classification problems with performance reported on well known data sets used in machine learning. The works are categorized according to the three primary methods of plasticity used-structural plasticity, weight plasticity, or plasticity of synaptic delays. Finally, we show the recent trend of confluence between concepts of deep learning and dendritic computations and highlight some future research directions.This article is part of a Special Issue entitled: SI: Dendritic contributions to biological and artificial computations. (c) 2022 IBRO. Published by Elsevier Ltd. All rights reserved.

Automated summary

In this paper, the nonlinear computational power provided by dendrites in biological and artificial neurons is discussed. Biological evidence, plasticity rules, and their impact on biological learning assessed by computational models are briefly presented. The computational implications include improved expressivity, efficient resource utilization, utilization of internal learning signals, and enabling continual learning. Examples of using dendritic computations to solve real-world classification problems are discussed, categorized based on the methods of plasticity used. The convergence between concepts of deep learning and dendritic computations is highlighted, along with future research directions.