Stochastic Computing and Spiking Neural Networks at SiPS’16

Professor Warren Gross recently presented a fascinating study investigating the relationship between stochastic computing and spiking neural networks:

Stochastic Computing Can Improve Upon Digital Spiking Neural Networks

This work, led by PhD student Sean C. Smithson, not only illustrates the parallels between digital spiking neural networks and stochastic computing, but also demonstrates that many computing elements in modern spiking hardware are, in fact, implementations of stochastic circuits. In addition, we show that stochastic computing design techniques can be leveraged in order to address shortcomings in current spiking neural network architectures.  This work was presented at the International Workshop on Signal Processing Systems (SiPS).