ICML 2021spotlight40 citations

Robust Representation Learning via Perceptual Similarity Metrics

Saeid A Taghanaki, Kristy Choi, Amir Hosein Khasahmadi, Anirudh Goyal

Abstract

A fundamental challenge in artificial intelligence is learning useful representations of data that yield good performance on a downstream classification task, without overfitting to spurious input features. Extracting such task-relevant predictive information becomes particularly difficult for noisy and high-dimensional real-world data. In this work, we propose Contrastive Input Morphing (CIM), a representation learning framework that learns input-space transformations of the data to mitigate the effect of irrelevant input features on downstream performance. Our method leverages a perceptual similarity metric via a triplet loss to ensure that the transformation preserves task-relevant information. Empirically, we demonstrate the efficacy of our approach on various tasks which typically suffer from the presence of spurious correlations: classification with nuisance information, out-of-distribution generalization, and preservation of subgroup accuracies. We additionally show that CIM is complementary to other mutual information-based representation learning techniques, and demonstrate that it improves the performance of variational information bottleneck (VIB) when used in conjunction.

BibTeX
@InProceedings{pmlr-v139-taghanaki21a,
  title = 	 {Robust Representation Learning via Perceptual Similarity Metrics},
  author =       {Taghanaki, Saeid A and Choi, Kristy and Khasahmadi, Amir Hosein and Goyal, Anirudh},
  booktitle = 	 {Proceedings of the 38th International Conference on Machine Learning},
  pages = 	 {10043--10053},
  year = 	 {2021},
  editor = 	 {Meila, Marina and Zhang, Tong},
  volume = 	 {139},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {18--24 Jul},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v139/taghanaki21a/taghanaki21a.pdf},
  url = 	 {https://proceedings.mlr.press/v139/taghanaki21a.html},
  abstract = 	 {A fundamental challenge in artificial intelligence is learning useful representations of data that yield good performance on a downstream classification task, without overfitting to spurious input features. Extracting such task-relevant predictive information becomes particularly difficult for noisy and high-dimensional real-world data. In this work, we propose Contrastive Input Morphing (CIM), a representation learning framework that learns input-space transformations of the data to mitigate the effect of irrelevant input features on downstream performance. Our method leverages a perceptual similarity metric via a triplet loss to ensure that the transformation preserves task-relevant information. Empirically, we demonstrate the efficacy of our approach on various tasks which typically suffer from the presence of spurious correlations: classification with nuisance information, out-of-distribution generalization, and preservation of subgroup accuracies. We additionally show that CIM is complementary to other mutual information-based representation learning techniques, and demonstrate that it improves the performance of variational information bottleneck (VIB) when used in conjunction.}
}
Robust Representation Learning via Perceptual Similarity Metrics · ICML 2021