Deep neural networks are biased towards simple functions.

Giacomo De Palma Bobak Toussi Kiani Seth Lloyd

Published in: CoRR (2018)

Keyphrases

neural network
pattern recognition
unsupervised learning
highly reliable
artificial neural networks
website
high quality
bayesian networks
case study
sufficient conditions
computer vision
back propagation
self organizing maps
multilayer perceptron
multi layer
cellular neural networks
data sets