Towards Repairing Neural Networks Correctly

TL;DR

This paper introduces a runtime verification approach that enhances neural network safety by strategically adding gates during execution, ensuring properties are satisfied without extensive static verification.

Contribution

It proposes a novel runtime verification method that combines static verification insights with dynamic corrections to improve neural network safety.

Findings

Effectively guarantees neural network safety properties.

Maintains consistency with original neural networks most of the time.

Demonstrates scalability over static verification methods.

Abstract

Neural networks are increasingly applied to support decision making in safety-critical applications (like autonomous cars, unmanned aerial vehicles and face recognition based authentication). While many impressive static verification techniques have been proposed to tackle the correctness problem of neural networks, it is possible that static verification may never be sufficiently scalable to handle real-world neural networks. In this work, we propose a runtime verification method to ensure the correctness of neural networks. Given a neural network and a desirable safety property, we adopt state-of-the-art static verification techniques to identify strategically locations to introduce additional gates which "correct" neural network behaviors at runtime. Experiment results show that our approach effectively generates neural networks which are guaranteed to satisfy the properties, whilst being consistent with the original neural network most of the time.