Adversarial Training Methods for Deep Learning: A Systematic Review

Abstract

Deep neural networks are exposed to the risk of adversarial attacks via the fast gradient sign method (FGSM), projected gradient descent (PGD) attacks, and other attack algorithms. Adversarial training is one of the methods used to defend against the threat of adversarial attacks. It is a training schema that utilizes an alternative objective function to provide model generalization for both adversarial data and clean data. In this systematic review, we focus particularly on adversarial training as a method of improving the defensive capacities and robustness of machine learning models. Specifically, we focus on adversarial sample accessibility through adversarial sample generation methods. The purpose of this systematic review is to survey state-of-the-art adversarial training and robust optimization methods to identify the research gaps within this field of applications. The literature search was conducted using Engineering Village (Engineering Village is an engineering literature search tool, which provides access to 14 engineering literature and patent databases), where we collected 238 related papers. The papers were filtered according to defined inclusion and exclusion criteria, and information was extracted from these papers according to a defined strategy. A total of 78 papers published between 2016 and 2021 were selected. Data were extracted and categorized using a defined strategy, and bar plots and comparison tables were used to show the data distribution. The findings of this review indicate that there are limitations to adversarial training methods and robust optimization. The most common problems are related to data generalization and overfitting.

Interpretable Multi-Scale Deep Learning for RNA Methylation Analysis across Multiple Species

“…Adversarial training is a technique used to improve the reliability and robustness of a model against intentional alterations in the data [ 55 ]. This is particularly important for the BERT model, which has a large number of parameters, as adversarial training helps prevent the model from overfitting the training data [ 56 ].…”

Resilience in the Context of Cyber Security: A Review of the Fundamental Concepts and Relevance

“…Adversarial training has proven to be an effective defense approach in training deep learning models to increase adversary robustness and can thus be improved and used to deal with attacks in general [69]. This is because, unlike conventional models, an additional step is added to the training of the model, where in addition to clean data, contradictory data is also used, increasing the robustness of the model against adversarial attacks [70,71]. Thus, the concept of adversarial training can also be used in cyber defense systems, increasing the resilience of these defense systems in the Monitoring-Detection, Resistance-Absorption, Response-Adaptation phases and especially in the Learning-Optimization phase.…”

Examining Adversarial Examples Impact on Artificial Intelligence Based Blockage Prediction Systems for Ultra-Wideband Communication in Indoor Environments

“…One of the mitigation techniques that researchers in the area of adversarial learning have proposed to protect the targeted model from adversarial example attacks is adversarial training [26]. This technique consists in introducing adversarial examples into input data to deceive the deep neural network (DNN) model [27].…”