Stochastic Neural Networks with Monotonic Activation Functions

Siamak Ravanbakhsh, Barnabás Póczos, Jeff G. Schneider, Dale Schuurmans, Russell Greiner

Abstract

We propose a Laplace approximation that creates a stochastic unit from any smooth monotonic activation function, using only Gaussian noise. This paper investigates the application of this stochastic approximation in training a family of Restricted Boltzmann Machines (RBM) that are closely linked to Bregman divergences. This family, that we call exponential family RBM (Exp-RBM), is a subset of the exponential family Harmoniums that expresses family members through a choice of smooth monotonic non-linearity for each neuron. Using contrastive divergence along with our Gaussian approximation, we show that Exp-RBM can learn useful representations using novel stochastic units.
BibTeX
@inproceedings{ravanbakhsh2016stochastic,
  author = {Siamak Ravanbakhsh and Barnabás Póczos and Jeff G. Schneider and Dale Schuurmans and Russell Greiner},
  title = {{Stochastic Neural Networks with Monotonic Activation Functions}},
  booktitle = {International Conference on Artificial Intelligence and Statistics},
  year = {2016},
  url = {https://openreview.net/forum?id=uF8QvT_ks_Y},
}
Figure