FAIR-Ensemble: When Fairness Naturally Emerges From Deep Ensembling
By Wei-Yin Ko, Daniel D'souza, Karina Nguyen, Randall Balestriero, Sara Hooker. 2023
16 citations recorded by the source corpus.
Ensembling multiple Deep Neural Networks (DNNs) is a simple and effective way to improve top-line metrics and to outperform a larger single model. In this work, we go beyond top-line metrics and instead explore the impact of ensembling on subgroup performances. Surprisingly, we observe that even with a simple homogeneous ensemble -- all the individual DNNs share the same training set, architecture, and design choices -- the minority group performance disproportionately improves with the number of models compared to the majority group, i.e. fairness naturally emerges from ensembling. Even more surprising, we find that this gain keeps occurring even when a large number of models is considered, e.g. $20$, despite the fact that the average performance of the ensemble plateaus with fewer models. Our work establishes that simple DNN ensembles can be a powerful tool for alleviating disparate impact from DNN classifiers, thus curbing algorithmic harm. We also explore why this is the case. We find that even in homogeneous ensembles, varying the sources of stochasticity through parameter initialization, mini-batch sampling, and data-augmentation realizations, results in different fairness outcomes.
Subfields: Evaluation & Benchmarks, Safety, Robustness & Interpretability
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Evaluation & Benchmarks
- Irene Solaiman
- Zeerak Talat
- Yacine Jernite
- Alexandra Sasha Luccioni
- Jesse Dodge
- Su Lin Blodgett
- Margaret Mitchell
- Sara Hooker
In Safety, Robustness & Interpretability
- Irene Solaiman
- Su Lin Blodgett
- Dylan Baker
- Zeerak Talat
- Marie-Therese Png
- Yacine Jernite
- William Agnew
- Alexandra Sasha Luccioni
Citations in the corpus
- Evaluating the Social Impact of Generative AI Systems in Systems and Society (2023, background): “These can include choices about many stages of the optimization process [237, 129]; privacy constraints [24], widely used compression techniques [109, 15, 169] and the choice hardware [273] have all been found to amplify harm on underrepresented protected attributes [28].”