Distributed Repair of Deep Neural Networks
Distributed Repair of Deep Neural Networks
复制标题
DOI：
10.1109/icst57152.2023.00017
复制
发表时间：
2023-04
期刊：
影响因子：
--
通讯作者：
Davide Li Calsi;Matias Duran;Xiaoyi Zhang;Paolo Arcaini;F. Ishikawa
中科院分区：
文献类型：
--
作者：
Davide Li Calsi;Matias Duran;Xiaoyi Zhang;Paolo Arcaini;F. Ishikawa
深度神经网络（DNN）被应用于多个安全关键领域，其可信度至关重要。例如，自动驾驶中用作分类器的DNN不应将检测到的物体误分类；然而，由于无法实现完美的准确率，应特别关注最关键的情况，比如行人。我们来自汽车领域的合作伙伴联盟已证实了这一点，他们为我们提供了不同误分类的具体风险等级。最近一种提升DNN性能的方法是定位导致误分类的DNN权重，然后对其进行调整（修复）以改善误分类情况。然而，当需要考虑多种误分类时，这些方法表现欠佳，并且它们没有考虑不同误分类的风险等级。为解决这一问题，我们提出了DISTRREP，一种分布式修复方法。该方法首先针对每种关键误分类找到最佳修复方案，然后通过考虑风险等级，将这些方案整合到单个修复后的DNN模型中。我们根据行业合作伙伴提出的要求，在三个DNN模型和一个自动驾驶图像数据集上对DISTRREP进行了评估。实验表明，DISTRREP比基于重新训练的基线方法以及其他不考虑风险的修复方法更为有效。
Deep Neural Networks (DNNs) are applied in several safety-critical domains and their trustworthiness is of paramount importance. For example, DNNs used in autonomous driving as classifiers should not misclassify detected objects; however, since obtaining perfect accuracy is not possible, special attention should be given to the most critical cases, e.g., pedestrians. This has been confirmed by the consortium of our partners from the automotive domain that provided us with specific risk levels for different misclassifications. A recent approach to improve DNN performance is to localise DNN weights responsible for the misclassifications and then adjust (repair) them to improve the misclassifications. However, they under-perform when they need to consider multiple misclassifications, and they do not consider the risk levels of the different misclassifications. To tackle this, we propose DISTRREP, a distributed repair approach that first finds the best fixes for each critical misclassification, and then integrates them in a single repaired DNN model, by considering the risk levels. We assess DISTRREP over three DNN models and a dataset of autonomous driving images, by considering requirements specified by our industrial partners. Experiments show that DISTRREP is more effective than baseline approaches based on retraining, and other risk-unaware repair approaches.