Robust Ensemble Model Training via Random Layer Sampling Against Adversarial Attack

TL;DR

This paper introduces a novel ensemble training method using random layer sampling to enhance deep neural network robustness against adversarial attacks, effectively hiding gradients and resisting gradient-based adversarial methods.

Contribution

The paper presents a new ensemble training framework with random layer sampling that improves adversarial robustness and gradient hiding in deep neural networks.

Findings

Enhanced robustness against adversarial attacks.

Effective gradient hiding through random layer sampling.

Improved performance demonstrated on three datasets.

Abstract

Deep neural networks have achieved substantial achievements in several computer vision areas, but have vulnerabilities that are often fooled by adversarial examples that are not recognized by humans. This is an important issue for security or medical applications. In this paper, we propose an ensemble model training framework with random layer sampling to improve the robustness of deep neural networks. In the proposed training framework, we generate various sampled model through the random layer sampling and update the weight of the sampled model. After the ensemble models are trained, it can hide the gradient efficiently and avoid the gradient-based attack by the random layer sampling method. To evaluate our proposed method, comprehensive and comparative experiments have been conducted on three datasets. Experimental results show that the proposed method improves the adversarial robustness.