Controlled Abstention Neural Networks for Identifying Skillful Predictions for Classification Problems

Elizabeth A. Barnes, Randal J. Barnes

Research output: Contribution to journalArticlepeer-review

3 Scopus citations


The earth system is exceedingly complex and often chaotic in nature, making prediction incredibly challenging: we cannot expect to make perfect predictions all of the time. Instead, we look for specific states of the system that lead to more predictable behavior than others, often termed “forecasts of opportunity.” When these opportunities are not present, scientists need prediction systems that are capable of saying “I don't know.” We introduce a novel loss function, termed the “NotWrong loss,” that allows neural networks to identify forecasts of opportunity for classification problems. The NotWrong loss introduces an abstention class that allows the network to identify the more confident samples and abstain (say “I don't know”) on the less confident samples. The abstention loss is designed to abstain on a user-defined fraction of the samples via a standard adaptive controller. Unlike many machine learning methods used to reject samples post-training, the NotWrong loss is applied during training to preferentially learn from the more confident samples. We show that the NotWrong loss outperforms other existing loss functions for multiple climate use cases. The implementation of the proposed loss function is straightforward in most network architectures designed for classification as it only requires the addition of an abstention class to the output layer and modification of the loss function.

Original languageEnglish (US)
Article numbere2021MS002573
JournalJournal of Advances in Modeling Earth Systems
Issue number12
StatePublished - Dec 2021

Bibliographical note

Funding Information:
The authors wish to thank the editor, David John Gagne, and an anonymous reviewer for helping us improve the paper. This work was funded, in part, by the NSF AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography (AI2ES) under NSF grant ICER‐2019758.

Publisher Copyright:
© 2021 The Authors. Journal of Advances in Modeling Earth Systems published by Wiley Periodicals LLC on behalf of American Geophysical Union.


  • classification
  • forecasts of opportunity
  • neural networks
  • prediction
  • uncertainty


Dive into the research topics of 'Controlled Abstention Neural Networks for Identifying Skillful Predictions for Classification Problems'. Together they form a unique fingerprint.

Cite this