Online Verification of Deep Neural Networks under Domain or Weight Shift

by   Tianhao Wei, et al.

Although neural networks are widely used, it remains challenging to formally verify the safety and robustness of neural networks in real-world applications. Existing methods are designed to verify the network before use, which is limited to relatively simple specifications and fixed networks. These methods are not ready to be applied to real-world problems with complex and/or dynamically changing specifications and networks. To effectively handle dynamically changing specifications and networks, the verification needs to be performed online when these changes take place. However, it is still challenging to run existing verification algorithms online. Our key insight is that we can leverage the temporal dependencies of these changes to accelerate the verification process, e.g., by warm starting new online verification using previous verified results. This paper establishes a novel framework for scalable online verification to solve real-world verification problems with dynamically changing specifications and/or networks, known as domain shift and weight shift respectively. We propose three types of techniques (branch management, perturbation tolerance analysis, and incremental computation) to accelerate the online verification of deep neural networks. Experiment results show that our online verification algorithm is up to two orders of magnitude faster than existing verification algorithms, and thus can scale to real-world applications.


page 1

page 2

page 3

page 4


Neuro-Symbolic Verification of Deep Neural Networks

Formal verification has emerged as a powerful approach to ensure the saf...

Voter Verification of BMD Ballots Is a Two-Part Question: Can They? Mostly, They Can. Do They? Mostly, They Don't

The question of whether or not voters actually verify ballots produced b...

Sound and Complete Verification of Polynomial Networks

Polynomial Networks (PNs) have demonstrated promising performance on fac...

Toward Reliable Neural Specifications

Having reliable specifications is an unavoidable challenge in achieving ...

Neural Network Robustness Verification on GPUs

Certifying the robustness of neural networks against adversarial attacks...

Towards Repairing Neural Networks Correctly

Neural networks are increasingly applied to support decision making in s...

veriFIRE: Verifying an Industrial, Learning-Based Wildfire Detection System

In this short paper, we present our ongoing work on the veriFIRE project...

Please sign up or login with your details

Forgot password? Click here to reset