Enhancing Representation Learning with Deep Classifiers in Presence of Shortcut

Amirhossein Ahmadian, Fredrik LindstenICASSP 2023 · OpenReview

Abstract

A deep neural classifier trained on an upstream task can be leveraged to boost the performance of another classifier in a related downstream task through the representations learned in hidden layers. However, presence of shortcuts (easy-to-learn features) in the upstream task can considerably impair the versatility of intermediate representations and, in turn, the downstream performance. In this paper, we propose a method to improve the representations learned by deep neural image classifiers in spite of a shortcut in upstream data. In our method, the upstream classification objective is augmented with a type of adversarial training where an auxiliary network, so called lens, fools the classifier by exploiting the shortcut in reconstructing images. Empirical comparisons in self-supervised and transfer learning problems with three shortcut-biased datasets suggest the advantages of our method in terms of downstream performance and/or training time.
BibTeX
@inproceedings{ahmadian2023enhancing,
  author = {Amirhossein Ahmadian and Fredrik Lindsten},
  title = {{Enhancing Representation Learning with Deep Classifiers in Presence of Shortcut}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2023},
  url = {https://openreview.net/forum?id=UVz5Q8HGyT},
}

Loading related content...

Figure