通过网络约简加速神经网络验证
Expediting Neural Network Verification via Network Reduction
AI总结：
本文提出验证前的网络约简预处理，通过消除稳定 ReLU 神经元并转换为 ReLU/Affine 序列化结构，显著提升 alpha-beta-crown、VeriNet 和 PRIMA 的验证速度与适用范围。
摘要：
人们已经提出了大量验证方法，用于验证深度神经网络的安全性质，以确保网络在关键应用中正确运行。然而，许多知名验证工具仍难以处理复杂的网络架构和庞大的网络规模。在本工作中，我们提出一种网络约简技术，将其作为验证之前的预处理方法。所提出的方法通过消除稳定的 ReLU 神经元来约简神经网络，并将其转换为由 ReLU 层和 Affine 层组成的序列化神经网络，这些层能够被大多数验证工具处理。我们将该约简技术实例化到最先进的完备与不完备验证工具上，包括 alpha-beta-crown、VeriNet 和 PRIMA。我们在大量基准测试上的实验表明，所提出的技术能够显著约简神经网络，并加快现有验证工具的速度。此外，实验结果还表明，网络约简通过将许多网络转换为序列化神经网络，能够提升现有验证工具在这些网络上的可用性。
英文摘要：
A wide range of verification methods have been proposed to verify the safety properties of deep neural networks ensuring that the networks function correctly in critical applications. However, many well-known verification tools still struggle with complicated network architectures and large network sizes. In this work, we propose a network reduction technique as a pre-processing method prior to verification. The proposed method reduces neural networks via eliminating stable ReLU neurons, and transforming them into a sequential neural network consisting of ReLU and Affine layers which can be handled by the most verification tools. We instantiate the reduction technique on the state-of-the-art complete and incomplete verification tools, including alpha-beta-crown, VeriNet and PRIMA. Our experiments on a large set of benchmarks indicate that the proposed technique can significantly reduce neural networks and speed up existing verification tools. Furthermore, the experiment results also show that network reduction can improve the availability of existing verification tools on many networks by reducing them into sequential neural networks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
