Toward Runtime-Throttleable Neural Networks

TL;DR

This paper introduces runtime-throttleable neural networks that adapt their resource use dynamically, enabling efficient deployment on resource-constrained edge devices with minimal accuracy loss.

Contribution

It proposes a generic block-level gating method to create neural networks that can be throttled at runtime, balancing performance and resource consumption.

Findings

Smooth performance throttling achieved across various operating points

Applicable to standard CNN architectures for image classification and detection

Minimal accuracy loss during dynamic resource adaptation

Abstract

As deep neural network (NN) methods have matured, there has been increasing interest in deploying NN solutions to "edge computing" platforms such as mobile phones or embedded controllers. These platforms are often resource-constrained, especially in energy storage and power, but state-of-the-art NN architectures are designed with little regard for resource use. Existing techniques for reducing the resource footprint of NN models produce static models that occupy a single point in the trade-space between performance and resource use. This paper presents an approach to creating runtime-throttleable NNs that can adaptively balance performance and resource use in response to a control signal. Throttleable networks allow intelligent resource management, for example by allocating fewer resources in "easy" conditions or when battery power is low. We describe a generic formulation of throttling via block-level gating, apply it to create throttleable versions of several standard CNN architectures, and demonstrate that our approach allows smooth performance throttling over a wide range of operating points in image classification and object detection tasks, with only a small loss in peak accuracy.