论文标题
一种验证卷积神经网络的抽象方法
An Abstraction-Refinement Approach to Verifying Convolutional Neural Networks
论文作者
论文摘要
卷积神经网络由于其在计算机视觉，图像处理等领域的出色表现而获得了广泛的知名度。不幸的是，现在众所周知，卷积网络通常会产生错误的结果 - 例如，这些网络输入的小扰动可能导致严重的分类错误。近年来，已经提出了许多验证方法来证明没有此类错误，但是这些错误通常适用于完全连接的网络，并且在应用于卷积网络时会遭受加重的可伸缩性问题。为了解决这一差距，我们在这里提出了CNN-ABS框架，该框架尤其针对卷积网络的验证。 CNN-ABS的核心是一种抽象 - 进行技术，它通过删除卷积连接的方式简化了验证问题，以极大地造成对原始问题的过度评价。如果结果问题过于抽象，则恢复这些连接。 CNN-ABS旨在使用现有验证引擎作为后端，我们的评估表明，它可以显着提高最先进的DNN验证引擎的性能，从而使运行时平均减少15.7％。
Convolutional neural networks have gained vast popularity due to their excellent performance in the fields of computer vision, image processing, and others. Unfortunately, it is now well known that convolutional networks often produce erroneous results - for example, minor perturbations of the inputs of these networks can result in severe classification errors. Numerous verification approaches have been proposed in recent years to prove the absence of such errors, but these are typically geared for fully connected networks and suffer from exacerbated scalability issues when applied to convolutional networks. To address this gap, we present here the Cnn-Abs framework, which is particularly aimed at the verification of convolutional networks. The core of Cnn-Abs is an abstraction-refinement technique, which simplifies the verification problem through the removal of convolutional connections in a way that soundly creates an over-approximation of the original problem; and which restores these connections if the resulting problem becomes too abstract. Cnn-Abs is designed to use existing verification engines as a backend, and our evaluation demonstrates that it can significantly boost the performance of a state-of-the-art DNN verification engine, reducing runtime by 15.7% on average.
