Continuous Sigmoidal Belief Networks Trained using Slice Sampling

Real-valued random hidden variables can be useful for modelling latent structure that explains correlations among observed variables. I propose a simple unit that adds zero-mean Gaussian noise to its input before passing it through a sigmoidal squashing function. Such units can produce a variety of useful behaviors, ranging from deterministic to binary stochastic to continuous stochastic. I show how \slice sampling" can be used for inference and learning in top-down networks of these units and demonstrate learning on two simple problems.

• Type: Conference
• Year: 1996
• Where: NIPS
• Authors: Brendan J. Frey
• Keywords: NIPS 1996, NIPS 2007, Observed Variables, Random Hidden Variables, Zero-mean Gaussian Noise