Adversarial Robustness Toolbox v1.3.1 releases: crafting and analysis of attacks and defense methods for machine learning models

Adversarial Robustness Toolbox

Adversarial Robustness 360 Toolbox (ART) is a Python library supporting developers and researchers in defending Machine Learning models (Deep Neural Networks, Gradient Boosted Decision Trees, Support Vector Machines, Random Forests, Logistic Regression, Gaussian Processes, Decision Trees, Scikit-learn Pipelines, etc.) against adversarial threats and helps to make AI systems more secure and trustworthy. Machine Learning models are vulnerable to adversarial examples, which are inputs (images, texts, tabular data, etc.) deliberately modified to produce a desired response by the Machine Learning model. ART provides the tools to build and deploy defenses and test them with adversarial attacks.

Defending Machine Learning models involves certifying and verifying model robustness and model hardening with approaches such as pre-processing inputs, augmenting training data with adversarial samples, and leveraging runtime detection methods to flag any inputs that might have been modified by an adversary. The attacks implemented in ART allow creating adversarial attacks against Machine Learning models which are required to test defenses with state-of-the-art threat models.

Supported attack and defense methods

The Adversarial Robustness Toolbox contains implementations of the following attacks:

• Deep Fool (Moosavi-Dezfooli et al., 2015)
• Fast Gradient Method (Goodfellow et al., 2014)
• Jacobian Saliency Map (Papernot et al., 2016)
• Universal Perturbation (Moosavi-Dezfooli et al., 2016)
• Virtual Adversarial Method (Moosavi-Dezfooli et al., 2015)
• C&W Attack (Carlini and Wagner, 2016)
• NewtonFool (Jang et al., 2017)

The following defense methods are also supported:

• Feature squeezing (Xu et al., 2017)
• Spatial smoothing (Xu et al., 2017)
• Label smoothing (Warde-Farley and Goodfellow, 2016)
• Adversarial training (Szegedy et al., 2013)
• Virtual adversarial training (Miyato et al., 2017)

The details of the work from IBM research can be found in the research paper. The ART toolbox is developed with the goal of helping developers better understand

• Measuring model robustness
• Model hardening
• Runtime detection

Changelog v1.3.1

Changed

• Changed the method fit of the deep-learning classifiers KerasClassifier, TensorFlowClassifier, TensorFlowV2Classifier, PyTorchClassifier, and MXClassifier in art.estimators.classification to support index labels in addition to one-hot-encoded labels. (#479)
• Changed the preprocessing defence art.defences.preprocessing.Mp3Compression to support input in format np.float32 in addition to np.int16 and updated related notebooks. (#482)

Fixed

• Fixed art.attacks.evasion.DeepFool to correctly apply the over-shoot step, previously the over-shoot vector was alwasy zero independent of epsilon. (#476)
• Fixed method set_params for attacks with multiple framework-specific implementations (art.attacks.evasion.AdversarialPatch, and art.attacks.evasion.ProjectedGradientDescent) to set attributes correctly and updated related notebooks, previously these set attributes would have been ignored by the attack. (#481)

Download && Tutorial

Copyright (C) IBM Corporation 2018