Evolutionary Multi-objective Architecture Search Framework: Application to COVID-19 3D CT Classification

TL;DR

This paper introduces EMARS, an evolutionary multi-objective architecture search framework for COVID-19 3D CT classification, which balances exploration and exploitation to find efficient, high-performing models with reduced search instability.

Contribution

It proposes a novel potential objective in NAS, improving model search stability and efficiency for COVID-19 CT classification tasks.

Findings

Models found by EMARS outperform prior methods on three datasets.

The potential objective reduces search time and improves model quality.

EMARS balances exploration and exploitation effectively.

Abstract

The COVID-19 pandemic has threatened global health. Many studies have applied deep convolutional neural networks (CNN) to recognize COVID-19 based on chest 3D computed tomography (CT). Recent works show that no model generalizes well across CT datasets from different countries, and manually designing models for specific datasets requires expertise; thus, neural architecture search (NAS) that aims to search models automatically has become an attractive solution. To reduce the search cost on large 3D CT datasets, most NAS-based works use the weight-sharing (WS) strategy to make all models share weights within a supernet; however, WS inevitably incurs search instability, leading to inaccurate model estimation. In this work, we propose an efficient Evolutionary Multi-objective ARchitecture Search (EMARS) framework. We propose a new objective, namely potential, which can help exploit promising models to indirectly reduce the number of models involved in weights training, thus alleviating search instability. We demonstrate that under objectives of accuracy and potential, EMARS can balance exploitation and exploration, i.e., reducing search time and finding better models. Our searched models are small and perform better than prior works on three public COVID-19 3D CT datasets.