Detection of Adversarial Training Examples in Poisoning Attacks through Anomaly Detection

Abstract

Machine learning has become an important component for many systems and applications including computer vision, spam filtering, malware and network intrusion detection, among others. Despite the capabilities of machine learning algorithms to extract valuable information from data and produce accurate predictions, it has been shown that these algorithms are vulnerable to attacks. Data poisoning is one of the most relevant security threats against machine learning systems, where attackers can subvert the learning process by injecting malicious samples in the training data. Recent work in adversarial machine learning has shown that the so-called optimal attack strategies can successfully poison linear classifiers, degrading the performance of the system dramatically after compromising a small fraction of the training dataset. In this paper we propose a defence mechanism to mitigate the effect of these optimal poisoning attacks based on outlier detection. We show empirically that the adversarial examples generated by these attack strategies are quite different from genuine points, as no detectability constrains are considered to craft the attack. Hence, they can be detected with an appropriate pre-filtering of the training dataset.

arXiv comment10 pages, 3 figures

References in corpus (5)

• Poisoning Attacks against Support Vector Machines
• Is feature selection secure against training data poisoning?
• Data Poisoning Attacks on Factorization-Based Collaborative Filtering
• Deceiving Google's Perspective API Built for Detecting Toxic Comments
• Security Analysis of Online Centroid Anomaly Detection

Cited by in corpus (27)

• Data Poisoning Attacks to Deep Learning Based Recommender Systems
• Security and Privacy Issues in Deep Learning
• Data Poisoning Attacks Against Federated Learning Systems
• Fawkes: Protecting Privacy against Unauthorized Deep Learning Models
• Poisoning Attacks with Generative Adversarial Nets
• Witches' Brew: Industrial Scale Data Poisoning via Gradient Matching
• On the Robustness of Random Forest Against Untargeted Data Poisoning: An Ensemble-Based Approach
• Explainable Data Poison Attacks on Human Emotion Evaluation Systems based on EEG Signals
• Defend Data Poisoning Attacks on Voice Authentication
• Rethinking the Backdoor Attacks' Triggers: A Frequency Perspective
• Subnet Replacement: Deployment-stage backdoor attack against deep neural networks in gray-box setting
• Defense Against Reward Poisoning Attacks in Reinforcement Learning
• Regularisation Can Mitigate Poisoning Attacks: A Novel Analysis Based on Multiobjective Bilevel Optimisation
• Ulixes: Facial Recognition Privacy with Adversarial Machine Learning
• Deep Probabilistic Models to Detect Data Poisoning Attacks
• Data Cleaning for Accurate, Fair, and Robust Models: A Big Data - AI Integration Approach
• Fine-grained Anomaly Detection via Multi-task Self-Supervision
• Regularization Can Help Mitigate Poisoning Attacks... with the Right Hyperparameters
• Enhancing the Antidote: Improved Pointwise Certifications against Poisoning Attacks
• An Investigation of Data Poisoning Defenses for Online Learning
• Robust Risk Minimization for Statistical Learning
• Classification Auto-Encoder based Detector against Diverse Data Poisoning Attacks
• Mixed Strategy Game Model Against Data Poisoning Attacks
• De-Pois: An Attack-Agnostic Defense against Data Poisoning Attacks
• Defending Against Adversarial Denial-of-Service Data Poisoning Attacks
• Defending SVMs against Poisoning Attacks: the Hardness and DBSCAN Approach
• Accumulative Poisoning Attacks on Real-time Data