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 (65)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Deep Learning with Differential Privacy
• 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
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Language Models are Few-Shot Learners
• Conditional Image Synthesis With Auxiliary Classifier GANs
• Conditional Image Generation with PixelCNN Decoders
• Equality of Opportunity in Supervised Learning
• Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
• DRAW: A Recurrent Neural Network For Image Generation
• Counterfactual Fairness
• CTRL: A Conditional Transformer Language Model for Controllable Generation
• Adversarial Feature Learning
• FakeCatcher: Detection of Synthetic Portrait Videos using Biological Signals
• Adversarial Attacks and Defences: A Survey
• C-RNN-GAN: Continuous recurrent neural networks with adversarial training
• Universal Transformers
• Generating Adversarial Malware Examples for Black-Box Attacks Based on GAN
• IDSGAN: Generative Adversarial Networks for Attack Generation against Intrusion Detection
• MaskGAN: Better Text Generation via Filling in the______
• Scalable Private Learning with PATE
• Differentially Private Generative Adversarial Network
• Release Strategies and the Social Impacts of Language Models
• PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples
• GANS for Sequences of Discrete Elements with the Gumbel-softmax Distribution
• Neural Voice Cloning with a Few Samples
• Understanding Membership Inferences on Well-Generalized Learning Models
• CycleGAN, a Master of Steganography
• MuseGAN: Multi-track Sequential Generative Adversarial Networks for Symbolic Music Generation and Accompaniment
• 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
• The Robust Manifold Defense: Adversarial Training using Generative Models
• Stolen Memories: Leveraging Model Memorization for Calibrated White-Box Membership Inference
• MelNet: A Generative Model for Audio in the Frequency Domain
• HYPE: A Benchmark for Human eYe Perceptual Evaluation of Generative Models
• Modeling Tabular data using Conditional GAN
• Black-Box Attacks against RNN based Malware Detection Algorithms
• Adversarial Audio Synthesis
• DeepFaceLab: Integrated, flexible and extensible face-swapping framework
• Generative Adversarial Trainer: Defense to Adversarial Perturbations with GAN
• Adversarial Text Generation via Feature-Mover's Distance
• On the Effects of Batch and Weight Normalization in Generative Adversarial Networks
• Spatial Image Steganography Based on Generative Adversarial Network
• Coverless Information Hiding Based on Generative adversarial networks
• Finger-GAN: Generating Realistic Fingerprint Images Using Connectivity Imposed GAN
• Label-Consistent Backdoor Attacks
• Generative Models for Effective ML on Private, Decentralized Datasets
• 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
• Skill Rating for Generative Models
• 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
• AAG-Stega: Automatic Audio Generation-based Steganography
• Featurized Bidirectional GAN: Adversarial Defense via Adversarially Learned Semantic Inference
• Statistics of Deep Generated Images
• Synthetic Data -- Anonymisation Groundhog Day
• Provably Secure Generative Linguistic Steganography
• Neural Linguistic Steganography
• DomainGAN: Generating Adversarial Examples to Attack Domain Generation Algorithm Classifiers