Enforcing Equity in Neural Climate Emulators

TL;DR

This paper introduces a novel loss function to bias neural climate emulators towards equitable predictions across different human development index groups, balancing accuracy and fairness.

Contribution

It proposes a flexible fairness-enforcing loss function for neural networks that can be tailored to various equity metrics in climate emulation tasks.

Findings

Emulators trained with the new loss function produce more equitable predictions.

Adjusting the fairness weight in the loss function controls the tradeoff between accuracy and equity.

Proper hyperparameter tuning minimizes performance loss while improving fairness.

Abstract

Neural network emulators have become an invaluable tool for a wide variety of climate and weather prediction tasks. While showing incredibly promising results, these networks do not have an inherent ability to produce equitable predictions. That is, they are not guaranteed to provide a uniform quality of prediction along any particular class or group of people. This potential for inequitable predictions motivates the need for explicit representations of fairness in these neural networks. To that end, we draw on methods for enforcing analytical physical constraints in neural networks to bias networks towards more equitable predictions. We demonstrate the promise of this methodology using the task of climate model emulation. Specifically, we propose a custom loss function which punishes emulators with unequal quality of predictions across any prespecified regions or category, here defined using human development index (HDI). This loss function weighs a standard loss metric such as mean squared error against another metric which captures inequity along the equity category (HDI), allowing us to adjust the priority of each term before training. Importantly, the loss function does not specify a particular definition of equity to bias the neural network towards, opening the door for custom fairness metrics. Our results show that neural climate emulators trained with our loss function provide more equitable predictions and that the equity metric improves with greater weighting in the loss function. We empirically demonstrate that while there is a tradeoff between accuracy and equity when prioritizing the latter during training, an appropriate selection of the equity priority hyperparameter can minimize loss of performance.