Mapping Generative Models onto a Network of Digital Spiking Neurons

TL;DR

This paper presents a novel method for implementing generative Restricted Boltzmann Machines on digital neuromorphic hardware, demonstrating a proof-of-concept for low-power, parallelized cognitive computing.

Contribution

It introduces a systematic approach to map RBMs onto neuromorphic systems using bio-inspired neurons and automates the design process for optimal resource utilization.

Findings

Successful mapping of RBM inference onto IBM TrueNorth hardware

Validation of generative performance metrics on neuromorphic platform

First implementation of generative RBM on neuromorphic VLSI

Abstract

Stochastic neural networks such as Restricted Boltzmann Machines (RBMs) have been successfully used in applications ranging from speech recognition to image classification. Inference and learning in these algorithms use a Markov Chain Monte Carlo procedure called Gibbs sampling, where a logistic function forms the kernel of this sampler. On the other side of the spectrum, neuromorphic systems have shown great promise for low-power and parallelized cognitive computing, but lack well-suited applications and automation procedures. In this work, we propose a systematic method for bridging the RBM algorithm and digital neuromorphic systems, with a generative pattern completion task as proof of concept. For this, we first propose a method of producing the Gibbs sampler using bio-inspired digital noisy integrate-and-fire neurons. Next, we describe the process of mapping generative RBMs trained offline onto the IBM TrueNorth neurosynaptic processor -- a low-power digital neuromorphic VLSI substrate. Mapping these algorithms onto neuromorphic hardware presents unique challenges in network connectivity and weight and bias quantization, which, in turn, require architectural and design strategies for the physical realization. Generative performance metrics are analyzed to validate the neuromorphic requirements and to best select the neuron parameters for the model. Lastly, we describe a design automation procedure which achieves optimal resource usage, accounting for the novel hardware adaptations. This work represents the first implementation of generative RBM inference on a neuromorphic VLSI substrate.