Feature Reconstruction From Outputs Can Mitigate Simplicity Bias in Neural Networks

Sravanti Addepalli, Anshul Nasery, Venkatesh Babu Radhakrishnan, Praneeth Netrapalli, Prateek Jain 0002. Feature Reconstruction From Outputs Can Mitigate Simplicity Bias in Neural Networks. In The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-5, 2023. OpenReview.net, 2023. [doi]

@inproceedings{AddepalliNRN023,
  title = {Feature Reconstruction From Outputs Can Mitigate Simplicity Bias in Neural Networks},
  author = {Sravanti Addepalli and Anshul Nasery and Venkatesh Babu Radhakrishnan and Praneeth Netrapalli and Prateek Jain 0002},
  year = {2023},
  url = {https://openreview.net/pdf?id=zH9GcZ3ZGXu},
  researchr = {https://researchr.org/publication/AddepalliNRN023},
  cites = {0},
  citedby = {0},
  booktitle = {The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-5, 2023},
  publisher = {OpenReview.net},
}