Adversarial attacks and defenses in explainable artificial intelligence: A survey

Abstract

Explainable artificial intelligence (XAI) methods are portrayed as a remedy for debugging and trusting statistical and deep learning models, as well as interpreting their predictions. However, recent advances in adversarial machine learning (AdvML) highlight the limitations and vulnerabilities of state-of-the-art explanation methods, putting their security and trustworthiness into question. The possibility of manipulating, fooling or fairwashing evidence of the model's reasoning has detrimental consequences when applied in high-stakes decision-making and knowledge discovery. This survey provides a comprehensive overview of research concerning adversarial attacks on explanations of machine learning models, as well as fairness metrics. We introduce a unified notation and taxonomy of methods facilitating a common ground for researchers and practitioners from the intersecting research fields of AdvML and XAI. We discuss how to defend against attacks and design robust interpretation methods. We contribute a list of existing insecurities in XAI and outline the emerging research directions in adversarial XAI (AdvXAI). Future work should address improving explanation methods and evaluation protocols to take into account the reported safety issues.

Accepted by Information Fusion

References in corpus (33)

• XGBoost: A Scalable Tree Boosting System
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
• One pixel attack for fooling deep neural networks
• All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneously
• Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
• BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain
• Evasion Attacks against Machine Learning at Test Time
• How to Explain Individual Classification Decisions
• From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI
• Towards Robust Interpretability with Self-Explaining Neural Networks
• A Comprehensive Taxonomy for Explainable Artificial Intelligence: A Systematic Survey of Surveys on Methods and Concepts
• Ground Truth Evaluation of Neural Network Explanations with CLEVR-XAI
• Adversarial Machine Learning in Image Classification: A Survey Towards the Defender's Perspective
• Wild Patterns Reloaded: A Survey of Machine Learning Security against Training Data Poisoning
• Quantus: An Explainable AI Toolkit for Responsible Evaluation of Neural Network Explanations and Beyond
• Model-agnostic Feature Importance and Effects with Dependent Features -- A Conditional Subgroup Approach
• S-LIME: Stabilized-LIME for Model Explanation
• Acquisition of Chess Knowledge in AlphaZero
• Counterfactual State Explanations for Reinforcement Learning Agents via Generative Deep Learning
• AI Certification: Advancing Ethical Practice by Reducing Information Asymmetries
• Checklist for responsible deep learning modeling of medical images based on COVID-19 detection studies
• Fairlearn: Assessing and Improving Fairness of AI Systems
• Adversarial Explanations for Understanding Image Classification Decisions and Improved Neural Network Robustness
• Towards Evaluating Explanations of Vision Transformers for Medical Imaging
• The Grammar of Interactive Explanatory Model Analysis
• A Step Toward More Inclusive People Annotations for Fairness
• The Bouncer Problem: Challenges to Remote Explainability
• Adversarial Inter-Group Link Injection Degrades the Fairness of Graph Neural Networks
• The four-fifths rule is not disparate impact: a woeful tale of epistemic trespassing in algorithmic fairness
• Fooling Partial Dependence via Data Poisoning
• secml: A Python Library for Secure and Explainable Machine Learning
• Foiling Explanations in Deep Neural Networks
• BMVC 2019: Workshop on Interpretable and Explainable Machine Vision

Cited by in corpus (6)

• Interpretable machine learning for time-to-event prediction in medicine and healthcare
• On the Robustness of Global Feature Effect Explanations
• When Can You Trust Your Explanations? A Robustness Analysis on Feature Importances
• Autonomous Cyber Resilience via a Co-Evolutionary Arms Race within a Fortified Digital Twin Sandbox
• Explainable Adversarial Attacks on Coarse-to-Fine Classifiers
• Birds look like cars: Adversarial analysis of intrinsically interpretable deep learning