Energy-Efficient Hybrid Stochastic-Binary Neural Networks for Near-Sensor Computing

TL;DR

This paper introduces a hybrid stochastic-binary neural network design for near-sensor computing that significantly improves energy efficiency while maintaining high accuracy, addressing power and bandwidth constraints.

Contribution

The paper proposes a novel stochastic-binary hybrid neural network architecture with new stochastic arithmetic units and retraining strategies for near-sensor applications.

Findings

Achieves 9.8x energy efficiency savings.

Maintains application accuracy within 0.05% of all-binary designs.

Uses new stochastic adders and multipliers for improved precision.

Abstract

Recent advances in neural networks (NNs) exhibit unprecedented success at transforming large, unstructured data streams into compact higher-level semantic information for tasks such as handwriting recognition, image classification, and speech recognition. Ideally, systems would employ near-sensor computation to execute these tasks at sensor endpoints to maximize data reduction and minimize data movement. However, near- sensor computing presents its own set of challenges such as operating power constraints, energy budgets, and communication bandwidth capacities. In this paper, we propose a stochastic- binary hybrid design which splits the computation between the stochastic and binary domains for near-sensor NN applications. In addition, our design uses a new stochastic adder and multiplier that are significantly more accurate than existing adders and multipliers. We also show that retraining the binary portion of the NN computation can compensate for precision losses introduced by shorter stochastic bit-streams, allowing faster run times at minimal accuracy losses. Our evaluation shows that our hybrid stochastic-binary design can achieve 9.8x energy efficiency savings, and application-level accuracies within 0.05% compared to conventional all-binary designs.