Minimising Contrastive Divergence in Noisy, Mixed-mode VLSI Neurons

Part of Advances in Neural Information Processing Systems 16 (NIPS 2003)

Authors

Hsin Chen, Patrice Fleury, Alan Murray

Abstract

This paper presents VLSI circuits with continuous-valued proba- bilistic behaviour realized by injecting noise into each computing unit(neuron). Interconnecting the noisy neurons forms a Contin- uous Restricted Boltzmann Machine (CRBM), which has shown promising performance in modelling and classifying noisy biomed- ical data. The Minimising-Contrastive-Divergence learning algo- rithm for CRBM is also implemented in mixed-mode VLSI, to adapt the noisy neurons’ parameters on-chip.