@COMMENT This file was generated by bib2html.pl version 0.90
@COMMENT written by Patrick Riley
@COMMENT This file came from Peter Stone's publication pages at
@COMMENT http://www.cs.utexas.edu/~pstone/papers
@TechReport{Decebal17,
author={Decebal Constantin Mocanu and Elena Mocanu and Peter Stone and Phuong H. Nguyen and Madeleine Gibescu and Antonio Liotta},
title={Evolutionary Training of Sparse Artificial Neural Networks: {A} Network Science Perspective},
year = "2017",
month = "August",
institution = "arXiv",
number = "arXiv e-Prints 1707.04780",
abstract="
Through the success of deep learning, Artificial Neural
Networks (ANNs) are among the most used artificial intelligence
methods nowadays. ANNs have led to major breakthroughs in
various domains, such as particle physics, reinforcement
learning, speech recognition, computer vision, and so
on. Taking inspiration from the network properties of
biological neural networks (e.g. sparsity, scale-freeness), we
argue that (contrary to general practice) Artificial Neural
Networks (ANN), too, should not have fully-connected layers. We
show how ANNs perform perfectly well with sparsely-connected
layers. Following a Darwinian evolutionary approach, we propose
a novel algorithm which evolves an initial random sparse
topology (i.e. an Erd\H{o}s-R\'enyi random graph) of two
consecutive layers of neurons into a scale-free topology,
during the ANN training process. The resulting sparse layers
can safely replace the corresponding fully-connected
layers. Our method allows to quadratically reduce the number of
parameters in the fully conencted layers of ANNs, yielding
quadratically faster computational times in both phases
(i.e. training and inference), at no decrease in accuracy. We
demonstrate our claims on two popular ANN types (restricted
Boltzmann machine and multi-layer perceptron), on two types of
tasks (supervised and unsupervised learning), and on 14
benchmark datasets. We anticipate that our approach will enable
ANNs having billions of neurons and evolved topologies to be
capable of handling complex real-world tasks that are
intractable using state-of-the-art methods. ",
wwwnote={Available online},
}