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.

The authors introduce a new neural network model which enables a client to verify that its request to execute inference over a neural network has been done correctly by a server. SafetyNets provide the client a high-probability guarantee of detecting wrong/lazy executions by the server, and incurs only a low overhead for the server while significantly decreasing the client's computation compared to executing the network itself. SafetyNets come with severe restrictions on the network during inference, but experimentally these restrictions are shown to have little impact on network performance. Quality: The experiments are convincing; the theory is an extension to the specific setting of neural networks of related work on interactive proofs. Clarity: This paper is very clear and easy to understand.

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.