Searching Toward Pareto-Optimal Device-Aware Neural Architectures

TL;DR

This paper reviews recent neural architecture search (NAS) methods and introduces multi-objective NAS approaches, MONAS and DPP-Net, that optimize for accuracy, latency, and energy across diverse devices, achieving Pareto optimality.

Contribution

It surveys NAS advancements and presents two multi-objective NAS methods that consider device-specific constraints, enabling optimized architectures for various hardware.

Findings

MONAS and DPP-Net find Pareto-optimal architectures for multiple device types.

These methods balance accuracy with latency and energy consumption.

Experimental results demonstrate effective multi-objective optimization across devices.

Abstract

Recent breakthroughs in Neural Architectural Search (NAS) have achieved state-of-the-art performance in many tasks such as image classification and language understanding. However, most existing works only optimize for model accuracy and largely ignore other important factors imposed by the underlying hardware and devices, such as latency and energy, when making inference. In this paper, we first introduce the problem of NAS and provide a survey on recent works. Then we deep dive into two recent advancements on extending NAS into multiple-objective frameworks: MONAS and DPP-Net. Both MONAS and DPP-Net are capable of optimizing accuracy and other objectives imposed by devices, searching for neural architectures that can be best deployed on a wide spectrum of devices: from embedded systems and mobile devices to workstations. Experimental results are poised to show that architectures found by MONAS and DPP-Net achieves Pareto optimality w.r.t the given objectives for various devices.