Generative Models for Security: Attacks, Defenses, and Opportunities

Abstract

Generative models learn the distribution of data from a sample dataset and can then generate new data instances. Recent advances in deep learning has brought forth improvements in generative model architectures, and some state-of-the-art models can (in some cases) produce outputs realistic enough to fool humans. We survey recent research at the intersection of security and privacy and generative models. In particular, we discuss the use of generative models in adversarial machine learning, in helping automate or enhance existing attacks, and as building blocks for defenses in contexts such as intrusion detection, biometrics spoofing, and malware obfuscation. We also describe the use of generative models in diverse applications such as fairness in machine learning, privacy-preserving data synthesis, and steganography. Finally, we discuss new threats due to generative models: the creation of synthetic media such as deepfakes that can be used for disinformation.

References in corpus (29)

• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• On the difficulty of training Recurrent Neural Networks
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• Language Models are Few-Shot Learners
• Equality of Opportunity in Supervised Learning
• Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
• DRAW: A Recurrent Neural Network For Image Generation
• C-RNN-GAN: Continuous recurrent neural networks with adversarial training
• Generating Adversarial Malware Examples for Black-Box Attacks Based on GAN
• GANS for Sequences of Discrete Elements with the Gumbel-softmax Distribution
• CycleGAN, a Master of Steganography
• Generative Poisoning Attack Method Against Neural Networks
• Weighted Transformer Network for Machine Translation
• Towards Deeper Understanding of Variational Autoencoding Models
• Do GANs actually learn the distribution? An empirical study
• MelNet: A Generative Model for Audio in the Frequency Domain
• Black-Box Attacks against RNN based Malware Detection Algorithms
• Coverless Information Hiding Based on Generative adversarial networks
• Finger-GAN: Generating Realistic Fingerprint Images Using Connectivity Imposed GAN
• Generative Models for Effective ML on Private, Decentralized Datasets
• Label-Consistent Backdoor Attacks
• Ensembles of Generative Adversarial Networks
• A Deep Learning-based Framework for Conducting Stealthy Attacks in Industrial Control Systems
• Differentially Private Synthetic Data: Applied Evaluations and Enhancements
• Improving Deep Learning with Differential Privacy using Gradient Encoding and Denoising
• Heard More Than Heard: An Audio Steganography Method Based on GAN
• Fréchet Audio Distance: A Metric for Evaluating Music Enhancement Algorithms
• Provably Secure Generative Linguistic Steganography
• DomainGAN: Generating Adversarial Examples to Attack Domain Generation Algorithm Classifiers