Verifying the Generalization of Deep Learning to Out-of-Distribution Domains

TL;DR

This paper presents a verification-based method to assess and improve the generalization of deep neural networks to unseen input domains, crucial for safety-critical and real-world applications.

Contribution

It introduces a novel approach that uses DNN verification tools to measure and verify the robustness of DNN decision rules across new, unencountered input domains.

Findings

Effective in identifying generalization capabilities of DNNs.

Applicable to both supervised and unsupervised benchmarks.

Demonstrated on a real-world deep reinforcement learning system.

Abstract

Deep neural networks (DNNs) play a crucial role in the field of machine learning, demonstrating state-of-the-art performance across various application domains. However, despite their success, DNN-based models may occasionally exhibit challenges with generalization, i.e., may fail to handle inputs that were not encountered during training. This limitation is a significant challenge when it comes to deploying deep learning for safety-critical tasks, as well as in real-world settings characterized by substantial variability. We introduce a novel approach for harnessing DNN verification technology to identify DNN-driven decision rules that exhibit robust generalization to previously unencountered input domains. Our method assesses generalization within an input domain by measuring the level of agreement between independently trained deep neural networks for inputs in this domain. We also efficiently realize our approach by using off-the-shelf DNN verification engines, and extensively evaluate it on both supervised and unsupervised DNN benchmarks, including a deep reinforcement learning (DRL) system for Internet congestion control -- demonstrating the applicability of our approach for real-world settings. Moreover, our research introduces a fresh objective for formal verification, offering the prospect of mitigating the challenges linked to deploying DNN-driven systems in real-world scenarios.