Stealing Hyperparameters in Machine Learning

Abstract

Hyperparameters are critical in machine learning, as different hyperparameters often result in models with significantly different performance. Hyperparameters may be deemed confidential because of their commercial value and the confidentiality of the proprietary algorithms that the learner uses to learn them. In this work, we propose attacks on stealing the hyperparameters that are learned by a learner. We call our attacks hyperparameter stealing attacks. Our attacks are applicable to a variety of popular machine learning algorithms such as ridge regression, logistic regression, support vector machine, and neural network. We evaluate the effectiveness of our attacks both theoretically and empirically. For instance, we evaluate our attacks on Amazon Machine Learning. Our results demonstrate that our attacks can accurately steal hyperparameters. We also study countermeasures. Our results highlight the need for new defenses against our hyperparameter stealing attacks for certain machine learning algorithms.

In the 39th IEEE Symposium on Security and Privacy (IEEE S & P), May 2018

References in corpus (3)

• Delving into Transferable Adversarial Examples and Black-box Attacks
• Poisoning Attacks against Support Vector Machines
• Data Poisoning Attacks on Factorization-Based Collaborative Filtering

Cited by in corpus (16)

• Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning
• When Machine Unlearning Jeopardizes Privacy
• Practical Blind Membership Inference Attack via Differential Comparisons
• Privacy-preserving Machine Learning through Data Obfuscation
• Security and Privacy Issues in Deep Learning
• Security Analysis of Deep Neural Networks Operating in the Presence of Cache Side-Channel Attacks
• Cache Telepathy: Leveraging Shared Resource Attacks to Learn DNN Architectures
• Attributing Fake Images to GANs: Learning and Analyzing GAN Fingerprints
• GAMIN: An Adversarial Approach to Black-Box Model Inversion
• Machine Learning with Membership Privacy using Adversarial Regularization
• Exploring Connections Between Active Learning and Model Extraction
• Auditing Data Provenance in Text-Generation Models
• Confidential Machine Learning Computation in Untrusted Environments: A Systems Security Perspective
• Robust Membership Encoding: Inference Attacks and Copyright Protection for Deep Learning
• CHEETAH: An Ultra-Fast, Approximation-Free, and Privacy-Preserved Neural Network Framework based on Joint Obscure Linear and Nonlinear Computations
• A Novel Privacy-Preserving Deep Learning Scheme without Using Cryptography Component