RC-DARTS: Resource Constrained Differentiable Architecture Search

TL;DR

RC-DARTS introduces a resource-constrained differentiable architecture search method that efficiently finds lightweight neural architectures with reduced size and complexity, maintaining high accuracy for image classification tasks.

Contribution

The paper presents a novel constrained optimization approach with a multi-level search strategy for resource-aware neural architecture search.

Findings

Achieves smaller, faster architectures with comparable accuracy.

Outperforms existing methods on CIFAR-10 and ImageNet datasets.

Effectively balances resource constraints with model performance.

Abstract

Recent advances show that Neural Architectural Search (NAS) method is able to find state-of-the-art image classification deep architectures. In this paper, we consider the one-shot NAS problem for resource constrained applications. This problem is of great interest because it is critical to choose different architectures according to task complexity when the resource is constrained. Previous techniques are either too slow for one-shot learning or does not take the resource constraint into consideration. In this paper, we propose the resource constrained differentiable architecture search (RC-DARTS) method to learn architectures that are significantly smaller and faster while achieving comparable accuracy. Specifically, we propose to formulate the RC-DARTS task as a constrained optimization problem by adding the resource constraint. An iterative projection method is proposed to solve the given constrained optimization problem. We also propose a multi-level search strategy to enable layers at different depths to adaptively learn different types of neural architectures. Through extensive experiments on the Cifar10 and ImageNet datasets, we show that the RC-DARTS method learns lightweight neural architectures which have smaller model size and lower computational complexity while achieving comparable or better performances than the state-of-the-art methods.