Nonparametric Weight Initialization of Neural Networks via Integral RepresentationDownload PDF

16 Apr 2024 (modified: 24 Dec 2013)ICLR 2014 conference submissionReaders: Everyone
Decision: submitted, no decision
Abstract: A new initialization method for hidden parameters in a neural network is proposed. Derived from the integral representation of the neural network, a nonparametric probability distribution of hidden parameters is introduced. In this proposal, hidden parameters are initialized by samples drawn from this distribution, and output parameters are fitted by ordinary linear regression. Numerical experiments show that backpropagation with proposed initialization converges faster than uniformly random initialization. Also it is shown that the proposed method achieves enough accuracy by itself without backpropagation in some cases.
7 Replies