DeepRepair: Style-Guided Repairing for Deep Neural Networks in the Real-World Operational Environment
DeepRepair: Style-Guided Repairing for Deep Neural Networks in the Real-World Operational Environment
复制标题
DOI：
10.1109/tr.2021.3096332
复制
发表时间：
2022-12
影响因子：
5.9
通讯作者：
Bing Yu;Hua Qi;Guo Qing;Felix Juefei-Xu;Xiaofei Xie;L. Ma;Jianjun Zhao
中科院分区：
文献类型：
--
作者：
Bing Yu;Hua Qi;Guo Qing;Felix Juefei-Xu;Xiaofei Xie;L. Ma;Jianjun Zhao
深度神经网络（DNN）由于其高性能而不断扩展到各个领域的应用。以及操作环境中的潜在未知噪声因素，例如天气，模糊，噪声等。因此，它对DNNS的现实世界应用构成了一个相当重要的问题：如何修复已部署的DNN用于校正下方的故障样本在本文中，我们可以收集有限的故障样本来损害其处理正常或清洁数据的能力，但我们提出了一种风格的引导数据增强，以在操作环境中修复DNN，以学习和介绍未知的失败此外，在训练数据中，我们进一步提出了基于聚类的失败数据，以实现更有效的样式引导数据增强，我们进行了大规模评估，以此进行模式。在现实世界中，与四种最先进的数据增强方法和两种DNN修复方法相比，我们的技术成功修复了三个卷积神经网络和两个复发性神经网络，平均为62.88％和39.02％图案分别比最先进的维修方法在最大的故障模式上获得更高的维修性能，并且在干净的数据集上的精度更高。
Deep neural networks (DNNs) are continuously expanding their application to various domains due to their high performance. Nevertheless, a well-trained DNN after deployment could oftentimes raise errors during practical use in the operational environment due to the mismatching between distributions of the training dataset and the potential unknown noise factors in the operational environment, e.g., weather, blur, noise, etc. Hence, it poses a rather important problem for the DNNs’ real-world applications: how to repair the deployed DNNs for correcting the failure samples under the deployed operational environment while not harming their capability of handling normal or clean data with limited failure samples we can collect. In this article, we propose a style-guided data augmentation for repairing DNN in the operational environment, which learns and introduces the unknown failure patterns within the failure samples into the training data via the style transfer. Moreover, we further propose the clustering-based failure data generation for much more effective style-guided data augmentation. We conduct a large-scale evaluation with 15 degradation factors that may happen in the real world and compare with four state-of-the-art data augmentation methods and two DNN repairing methods. Our technique successfully repairs three convolutional neural networks and two recurrent neural networks with averaging 62.88% and 39.02% accuracy enhancements on the 15 failure patterns, respectively, achieving higher repairing performance than state-of-the-art repairing methods on the most failure patterns with even better accuracy on clean datasets.