A Review of Neuroscience-Inspired Machine Learning

arXiv:2403.18929v1 Announce Type: cross
Abstract: One major criticism of deep learning centers around the biological implausibility of the credit assignment schema used for learning -- backpropagation of errors. This implausibility translates into practical limitations, spanning scientific fields, including incompatibility with hardware and non-differentiable implementations, thus leading to expensive energy requirements. In contrast, biologically plausible credit assignment is compatible with practically any learning condition and is energy-efficient. As a result, it accommodates hardware and scientific modeling, e.g. learning with physical systems …

abstract arxiv backpropagation contrast credit cs.lg cs.ne deep learning differentiable energy errors fields hardware limitations machine machine learning major neuroscience practical requirements review schema scientific type