Repairing Failure-inducing Inputs with Input Reflection

Yan Xiao, Yun Lin, Ivan Beschastnikh, Changsheng Sun, David S. Rosenblum, Jin Song Dong

Abstract

Trained with a sufficiently large training and testing dataset, Deep Neural Networks (DNNs) are expected to generalize. However, inputs may deviate from the training dataset distribution in real deployments. This is a fundamental issue with using a finite dataset, which may lead deployed DNNs to mis-predict in production.

BibTeX

@inproceedings{Xiao-al:ASE22,
  author    = {Yan Xiao and
               Yun Lin and
               Ivan Beschastnikh and
               Changsheng Sun and
               David S. Rosenblum and
               Jin Song Dong},
  title     = {Repairing Failure-inducing Inputs with Input Reflection},
  booktitle = {ASE},
  pages     = {85:1--85:13},
  publisher = {{ACM}},
  year      = {2022},
}

Repairing Failure-inducing Inputs with Input Reflection

Abstract

Related papers