Nearest Neighbor Representations of Neural Circuits

arXiv:2402.08751v1 Announce Type: cross
Abstract: Neural networks successfully capture the computational power of the human brain for many tasks. Similarly inspired by the brain architecture, Nearest Neighbor (NN) representations is a novel approach of computation. We establish a firmer correspondence between NN representations and neural networks. Although it was known how to represent a single neuron using NN representations, there were no results even for small depth neural networks. Specifically, for depth-2 threshold circuits, we provide explicit constructions for their …

abstract architecture arxiv brain computation computational cs.cc cs.dm cs.lg cs.ne human networks neural networks novel power tasks type