SafetyNets: Verifiable Execution of Deep Neural Networks on an Untrusted Cloud

Abstract

Inference using deep neural networks is often outsourced to the cloud since it is a computationally demanding task. However, this raises a fundamental issue of trust. How can a client be sure that the cloud has performed inference correctly? A lazy cloud provider might use a simpler but less accurate model to reduce its own computational load, or worse, maliciously modify the inference results sent to the client. We propose SafetyNets, a framework that enables an untrusted server (the cloud) to provide a client with a short mathematical proof of the correctness of inference tasks that they perform on behalf of the client. Specifically, SafetyNets develops and implements a specialized interactive proof (IP) protocol for verifiable execution of a class of deep neural networks, i.e., those that can be represented as arithmetic circuits. Our empirical results on three- and four-layer deep neural networks demonstrate the run-time costs of SafetyNets for both the client and server are low. SafetyNets detects any incorrect computations of the neural network by the untrusted server with high probability, while achieving state-of-the-art accuracy on the MNIST digit recognition (99.4%) and TIMIT speech recognition tasks (75.22%).

References in corpus (2)

• On the Computational Efficiency of Training Neural Networks
• Globally Optimal Training of Generalized Polynomial Neural Networks with Nonlinear Spectral Methods

Cited by in corpus (11)

• Security for Machine Learning-based Systems: Attacks and Challenges during Training and Inference
• CryptoNAS: Private Inference on a ReLU Budget
• On Polynomial Approximations for Privacy-Preserving and Verifiable ReLU Networks
• VeriML: Enabling Integrity Assurances and Fair Payments for Machine Learning as a Service
• SoK: Machine Learning Governance
• Outsourcing Private Machine Learning via Lightweight Secure Arithmetic Computation
• Agatha: Smart Contract for DNN Computation
• Technologies for Trustworthy Machine Learning: A Survey in a Socio-Technical Context
• ShadowNet: A Secure and Efficient On-device Model Inference System for Convolutional Neural Networks
• VeriDL: Integrity Verification of Outsourced Deep Learning Services (Extended Version)
• CheckNet: Secure Inference on Untrusted Devices