February 21, 2023
Over the past few years, deep neural networks (DNNs) have achieved tremendous success and have been continuously applied in many application domains. However, during the practical deployment in industrial tasks, DNNs are found to be erroneous-prone due to various reasons such as overfitting and lacking of robustness to real-world corruptions during practical usage. To address these challenges, many recent attempts have been made to repair DNNs for version updates under practical operational contexts by updating weights (i.e., network parameters) through retraining, fine-tuning, or direct weight fixing at a neural level. Nevertheless, existing solutions often neglect the effects of neural network architecture and weight relationships across neurons and layers. In this work, as the first attempt, we initiate to repair DNNs by jointly optimizing the architecture and weights at a higher (i.e., block) level. We first perform empirical studies to investigate the limitation of whole network-level and layer-level repairing, which motivates us to explore a novel repairing direction for DNN repair at the block level. To this end, we need to further consider techniques to address two key technical challenges, i.e., block localization, where we should localize the targeted block thatwe need to fix; and howto perform joint architecture and weight repairing. Specifically, we first propose adversarial-aware spectrum analysis for vulnerable block localization that considers the neurons’ status and weights’ gradients in blocks during the forward and backward processes, which enables more accurate candidate block localization for repairing even under a few examples. Then, we further propose the architecture-oriented search-based repairing that relaxes the targeted block to a continuous repairing search space at higher deep feature levels. By jointly optimizing the architecture and weights in that space, we can identify a much better block architecture. We implement our proposed repairing techniques as a tool, named ArchRepair, and conduct extensive experiments to validate the proposed method. The results show that our method can not only repair but also enhance accuracy & robustness, outperforming the state-of-the-art DNN repair techniques.
Written by
Felix Xu
Fuyuan Zhang
Hua Qi
Jianjun Zhao
Jianlang Chen
Lei Ma
Qing Guo
Zhijie Wang
Publisher
ACM Transactions on Software Engineering and Methodology (TOSEM)
March 09, 2023
Bilal Porgali, Vítor Albiero, Jordan Ryda, Cristian Canton Ferrer, Caner Hazirbas
March 09, 2023
February 28, 2023
Quentin Garrido, Adrien Bardes, Yann LeCun, Yubei Chen, Laurent Najman
February 28, 2023
January 10, 2023
Gokhan Uzunbas, Joena Zhang, Sara Cao, Ser-Nam Lim, Taipeng Tian, Bohyung Han, Seonguk Seo
January 10, 2023
January 04, 2023
Xi Liu, Panganamala Kumar, Ruida Zhou, Tao Liu
January 04, 2023