FACET: Fairness in Computer Vision Evaluation Benchmark

Michael Lohaus Chris Russell Matthäus Kleindessner + 4 lainnya

9 Maret 2022

Algorithmic fairness is frequently motivated in terms of a trade-off in which overall performance is decreased so as to improve performance on disadvantaged groups where the algorithm would otherwise be less accurate. Contrary to this, we find that applying existing fairness approaches to computer vision improve fairness by degrading the performance of classifiers across all groups (with increased degradation on the best performing groups). Extending the bias-variance decomposition for classification to fairness, we theoretically explain why the majority of fairness methods designed for low capacity models should not be used in settings involving high-capacity models, a scenario common to computer vision. We corroborate this analysis with extensive experimental support that shows that many of the fairness heuristics used in computer vision also degrade performance on the most disadvantaged groups. Building on these insights, we propose an adaptive augmentation strategy that, uniquely, of all methods tested, improves performance for the disadvantaged groups.

Changjian Shui Raghav Mehta T. Arbel

6 Maret 2023

Although deep learning (DL) models have shown great success in many medical image analysis tasks, deployment of the resulting models into real clinical contexts requires: (1) that they exhibit robustness and fairness across different sub-populations, and (2) that the confidence in DL model predictions be accurately expressed in the form of uncertainties. Unfortunately, recent studies have indeed shown significant biases in DL models across demographic subgroups (e.g., race, sex, age) in the context of medical image analysis, indicating a lack of fairness in the models. Although several methods have been proposed in the ML literature to mitigate a lack of fairness in DL models, they focus entirely on the absolute performance between groups without considering their effect on uncertainty estimation. In this work, we present the first exploration of the effect of popular fairness models on overcoming biases across subgroups in medical image analysis in terms of bottom-line performance, and their effects on uncertainty quantification. We perform extensive experiments on three different clinically relevant tasks: (i) skin lesion classification, (ii) brain tumour segmentation, and (iii) Alzheimer's disease clinical score regression. Our results indicate that popular ML methods, such as data-balancing and distributionally robust optimization, succeed in mitigating fairness issues in terms of the model performances for some of the tasks. However, this can come at the cost of poor uncertainty estimates associated with the model predictions. This tradeoff must be mitigated if fairness models are to be adopted in medical image analysis.

M. Scheuerman Kandrea Wade Caitlin Lustig + 1 lainnya

28 Mei 2020

Race and gender have long sociopolitical histories of classification in technical infrastructures-from the passport to social media. Facial analysis technologies are particularly pertinent to understanding how identity is operationalized in new technical systems. What facial analysis technologies can do is determined by the data available to train and evaluate them with. In this study, we specifically focus on this data by examining how race and gender are defined and annotated in image databases used for facial analysis. We found that the majority of image databases rarely contain underlying source material for how those identities are defined. Further, when they are annotated with race and gender information, database authors rarely describe the process of annotation. Instead, classifications of race and gender are portrayed as insignificant, indisputable, and apolitical. We discuss the limitations of these approaches given the sociohistorical nature of race and gender. We posit that the lack of critical engagement with this nature renders databases opaque and less trustworthy. We conclude by encouraging database authors to address both the histories of classification inherently embedded into race and gender, as well as their positionality in embedding such classifications.

Zhenpeng Chen M. Harman J Zhang + 2 lainnya

15 Juli 2022

This article provides a comprehensive survey of bias mitigation methods for achieving fairness in Machine Learning (ML) models. We collect a total of 341 publications concerning bias mitigation for ML classifiers. These methods can be distinguished based on their intervention procedure (i.e., pre-processing, in-processing, post-processing) and the technique they apply. We investigate how existing bias mitigation methods are evaluated in the literature. In particular, we consider datasets, metrics, and benchmarking. Based on the gathered insights (e.g., What is the most popular fairness metric? How many datasets are used for evaluating bias mitigation methods?), we hope to support practitioners in making informed choices when developing and evaluating new bias mitigation methods.

C. Haas Simon Caton

4 Oktober 2020

When Machine Learning technologies are used in contexts that affect citizens, companies as well as researchers need to be confident that there will not be any unexpected social implications, such as bias towards gender, ethnicity, and/or people with disabilities. There is significant literature on approaches to mitigate bias and promote fairness, yet the area is complex and hard to penetrate for newcomers to the domain. This article seeks to provide an overview of the different schools of thought and approaches that aim to increase the fairness of Machine Learning. It organizes approaches into the widely accepted framework of pre-processing, in-processing, and post-processing methods, subcategorizing into a further 11 method areas. Although much of the literature emphasizes binary classification, a discussion of fairness in regression, recommender systems, and unsupervised learning is also provided along with a selection of currently available open source libraries. The article concludes by summarizing open challenges articulated as five dilemmas for fairness research.