Explaining Vulnerabilities to Adversarial Machine Learning through Visual Analytics

Abstract

Machine learning models are currently being deployed in a variety of real-world applications where model predictions are used to make decisions about healthcare, bank loans, and numerous other critical tasks. As the deployment of artificial intelligence technologies becomes ubiquitous, it is unsurprising that adversaries have begun developing methods to manipulate machine learning models to their advantage. While the visual analytics community has developed methods for opening the black box of machine learning models, little work has focused on helping the user understand their model vulnerabilities in the context of adversarial attacks. In this paper, we present a visual analytics framework for explaining and exploring model vulnerabilities to adversarial attacks. Our framework employs a multi-faceted visualization scheme designed to support the analysis of data poisoning attacks from the perspective of models, data instances, features, and local structures. We demonstrate our framework through two case studies on binary classifiers and illustrate model vulnerabilities with respect to varying attack strategies.

IEEE VAST (Transactions on Visualization and Computer Graphics), 2019

References in corpus (13)

• Scikit-learn: Machine Learning in Python
• Explaining and Harnessing Adversarial Examples
• Understanding Neural Networks Through Deep Visualization
• Wild Patterns: Ten Years After the Rise of Adversarial Machine Learning
• Evasion Attacks against Machine Learning at Test Time
• Poisoning Attacks against Support Vector Machines
• Security Evaluation of Pattern Classifiers under Attack
• RetainVis: Visual Analytics with Interpretable and Interactive Recurrent Neural Networks on Electronic Medical Records
• Manifold: A Model-Agnostic Framework for Interpretation and Diagnosis of Machine Learning Models
• GAN Lab: Understanding Complex Deep Generative Models using Interactive Visual Experimentation
• ShapeShifter: Robust Physical Adversarial Attack on Faster R-CNN Object Detector
• The State of the Art in Integrating Machine Learning into Visual Analytics
• Analyzing the Noise Robustness of Deep Neural Networks

Cited by in corpus (18)

• The State of the Art in Enhancing Trust in Machine Learning Models with the Use of Visualizations
• StackGenVis: Alignment of Data, Algorithms, and Models for Stacking Ensemble Learning Using Performance Metrics
• Diagnosing Concept Drift with Visual Analytics
• FeatureEnVi: Visual Analytics for Feature Engineering Using Stepwise Selection and Semi-Automatic Extraction Approaches
• VisEvol: Visual Analytics to Support Hyperparameter Search through Evolutionary Optimization
• Class-constrained t-SNE: Combining Data Features and Class Probabilities
• A Survey of Visual Analytics Techniques for Machine Learning
• Should We Trust (X)AI? Design Dimensions for Structured Experimental Evaluations
• ManiVault: A Flexible and Extensible Visual Analytics Framework for High-Dimensional Data
• JailbreakLens: Visual Analysis of Jailbreak Attacks Against Large Language Models
• HardVis: Visual Analytics to Handle Instance Hardness Using Undersampling and Oversampling Techniques
• Visual Analytics of Neuron Vulnerability to Adversarial Attacks on Convolutional Neural Networks
• Revealing Neural Network Bias to Non-Experts Through Interactive Counterfactual Examples
• TrafPS: A Shapley-based Visual Analytics Approach to Interpret Traffic
• DeforestVis: Behavior Analysis of Machine Learning Models with Surrogate Decision Stumps
• A Review of Explainable Artificial Intelligence in Manufacturing
• A Visual Analytics Framework for Explaining and Diagnosing Transfer Learning Processes
• Explainable Adversarial Attacks in Deep Neural Networks Using Activation Profiles