Controlled abstention neural networks for identifying skillful predictions for classification problems

TL;DR

This paper introduces the NotWrong loss function, enabling neural networks to abstain from uncertain predictions during training, thereby improving classification performance in complex climate prediction tasks.

Contribution

The NotWrong loss function is a novel training method that incorporates abstention directly into neural networks for better confidence-based predictions.

Findings

Outperforms existing loss functions in climate classification tasks

Enables neural networks to abstain on uncertain samples during training

Simple to implement with minimal modifications to existing architectures

Abstract

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 PID 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.