Machine Learning models have been deployed across many different aspects of society, often in situations that affect social welfare. Although these models offer streamlined solutions to large problems, they may contain biases and treat groups or individuals unfairly based on protected attributes such as gender. In this paper, we introduce several examples of machine learning gender bias in practice followed by formalizations of fairness. We provide a survey of fairness research by detailing influential pre-processing, in-processing, and post-processing bias mitigation algorithms. We then propose an end-to-end bias mitigation framework, which employs a fusion of pre-, in-, and post-processing methods to leverage the strengths of each individual technique. We test this method, along with the standard techniques we review, on a deep neural network to analyze bias mitigation in a deep learning setting. We find that our end-to-end bias mitigation framework outperforms the baselines with respect to several fairness metrics, suggesting its promise as a method for improving fairness. As society increasingly relies on artificial intelligence to help in decision-making, addressing gender biases present in deep learning models is imperative. To provide readers with the tools to assess the fairness of machine learning models and mitigate the biases present in them, we discuss multiple open source packages for fairness in AI.
Beijing Academy of Artificial Intelligence（北京人工智能研究院）；The University of Sheffield（谢菲尔德大学）；University of Michigan（密歇根大学）；University of Colorado Boulder（科罗拉多大学博尔德分校）；Sichuan University（四川大学）；McGill University（麦吉尔大学）
As natural language processing (NLP) for gender bias becomes a significant interdisciplinary topic, the prevalent data-driven techniques such as large-scale language models suffer from data inadequacy and biased corpus, especially for languages with insufficient resources such as Chinese. To this end, we propose a Chinese cOrpus foR Gender bIas Probing and Mitigation CORGI-PM, which contains 32.9k sentences with high-quality labels derived by following an annotation scheme specifically developed for gender bias in the Chinese context. Moreover, we address three challenges for automatic textual…
University of Virginia（弗吉尼亚大学）；Salesforce Research（赛富时研究院）
Word embeddings derived from human-generated corpora inherit strong gender bias which can be further amplified by downstream models. Some commonly adopted debiasing approaches, including the seminal Hard Debias algorithm, apply post-processing procedures that project pre-trained word embeddings into a subspace orthogonal to an inferred gender subspace. We discover that semantic-agnostic corpus regularities such as word frequency captured by the word embeddings negatively impact the performance of these algorithms. We propose a simple but effective technique, Double Hard Debias, which purifies …
Published studies have suggested the bias of automated face-based gender classification algorithms across gender-race groups. Specifically, unequal accuracy rates were obtained for women and dark-skinned people. To mitigate the bias of gender classifiers, the vision community has developed several strategies. However, the efficacy of these mitigation strategies is demonstrated for a limited number of races mostly, Caucasian and African-American. Further, these strategies often offer a trade-off between bias and classification accuracy. To further advance the state-of-the-art, we leverage the p…
Visa Inc.（Visa公司）；Harvard John A. Paulson School Of Engineering And Applied Sciences（哈佛大学约翰·A·保尔森工程与应用科学学院）；FM: Systems（FM: Systems公司）；Autodesk（Autodesk公司）
Gender classification algorithms have important applications in many domains today such as demographic research, law enforcement, as well as human-computer interaction. Recent research showed that algorithms trained on biased benchmark databases could result in algorithmic bias. However, to date, little research has been carried out on gender classification algorithms' bias towards gender minorities subgroups, such as the LGBTQ and the non-binary population, who have distinct characteristics in gender expression. In this paper, we began by conducting surveys on existing benchmark databases for…
Explanatory Interactive Machine Learning for Bias Mitigation in Visual Gender Classification
Johannes Kepler University Linz（约翰内斯·开普勒大学林茨）；Institute of Creative Media Technologies（创意媒体技术研究所）；St. Pölten University of Applied Sciences（施蒂利克大学应用科学学院）；Institute of Computational Perception（计算感知研究所）
McGill University（麦吉尔大学）；Microsoft Research Montreal（微软研究院蒙特利尔分院）
We introduce a new benchmark for coreference resolution and NLI, Knowref, that targets common-sense understanding and world knowledge. Previous coreference resolution tasks can largely be solved by exploiting the number and gender of the antecedents, or have been handcrafted and do not reflect the diversity of naturally occurring text. We present a corpus of over 8,000 annotated text passages with ambiguous pronominal anaphora. These instances are both challenging and realistic. We show that various coreference systems, whether rule-based, feature-rich, or neural, perform significantly worse o…
Small business classification is a difficult and important task within many applications, including customer segmentation. Training on small business names introduces gender and geographic origin biases. A model for predicting one of 66 business types based only upon the business name was developed in this work (top-1 f1-score = 60.2%). Two approaches to removing the bias from this model are explored: replacing given names with a placeholder token, and augmenting the training data with gender-swapped examples. The results for these approaches is reported, and the bias in the model was reduced …
Manipal Institute of Technology（马尼帕尔理工学院）；MAHE（马尼帕尔高等教育学院）
With language models being deployed increasingly in the real world, it is essential to address the issue of the fairness of their outputs. The word embedding representations of these language models often implicitly draw unwanted associations that form a social bias within the model. The nature of gendered languages like Hindi, poses an additional problem to the quantification and mitigation of bias, owing to the change in the form of the words in the sentence, based on the gender of the subject. Additionally, there is sparse work done in the realm of measuring and debiasing systems for Indic …
A major problem with using automated classification systems is that if they are not engineered correctly and with fairness considerations, they could be detrimental to certain populations. Furthermore, while engineers have developed cutting-edge technologies for image classification, there is still a gap in the application of these models in human heritage collections, where data sets usually consist of low-quality pictures of people with diverse ethnicity, gender, and age. In this work, we evaluate three bias mitigation techniques using two state-of-the-art neural networks, Xception and Effic…
Towards Trustworthy Audio Deepfake Detection: A Systematic Framework for Diagnosing and Mitigating Gender Bias
School of Computing, Wichita State University（维克托利亚州立大学计算机学院）；Institut national de la recherche scientifique (INRS-EMT)（国家科学研究机构（INRS-EMT））；INRS-UQO Mixed Research Unit on Cybersecurity（INRS-UQO网络安全混合研究单位）
Comparative assessment of fairness definitions and bias mitigation strategies in machine learning-based diagnosis of Alzheimer's disease from MR images
National Technical University of Athens（雅典国家技术大学）；University of Pennsylvania（宾夕法尼亚大学）
This paper proposes two intuitive metrics, skew and stereotype, that quantify and analyse the gender bias present in contextual language models when tackling the WinoBias pronoun resolution task. We find evidence that gender stereotype correlates approximately negatively with gender skew in out-of-the-box models, suggesting that there is a trade-off between these two forms of bias. We investigate two methods to mitigate bias. The first approach is an online method which is effective at removing skew at the expense of stereotype. The second, inspired by previous work on ELMo, involves the fine-…
Utrecht University（乌得勒支大学）；University Medical Center Utrecht（乌得勒支大学医学中心）；Leiden University Medical Center（莱顿大学医学中心）；Leiden University（莱顿大学）
Fairness and bias are crucial concepts in artificial intelligence, yet they are relatively ignored in machine learning applications in clinical psychiatry. We computed fairness metrics and present bias mitigation strategies using a model trained on clinical mental health data. We collected structured data related to the admission, diagnosis, and treatment of patients in the psychiatry department of the University Medical Center Utrecht. We trained a machine learning model to predict future administrations of benzodiazepines on the basis of past data. We found that gender plays an unexpected ro…
Tokyo Metropolitan University（东京都立大学）；University of Liverpool（利物浦大学）
Word embeddings learnt from massive text collections have demonstrated significant levels of discriminative biases such as gender, racial or ethnic biases, which in turn bias the down-stream NLP applications that use those word embeddings. Taking gender-bias as a working example, we propose a debiasing method that preserves non-discriminative gender-related information, while removing stereotypical discriminative gender biases from pre-trained word embeddings. Specifically, we consider four types of information: \emph{feminine}, \emph{masculine}, \emph{gender-neutral} and \emph{stereotypical},…
Artificial Intelligence has the capacity to amplify and perpetuate societal biases and presents profound ethical implications for society. Gender bias has been identified in the context of employment advertising and recruitment tools, due to their reliance on underlying language processing and recommendation algorithms. Attempts to address such issues have involved testing learned associations, integrating concepts of fairness to machine learning and performing more rigorous analysis of training data. Mitigating bias when algorithms are trained on textual data is particularly challenging given…
Tsinghua University（清华大学）；University of California, Los Angeles（加州大学洛杉矶分校）
Advanced machine learning techniques have boosted the performance of natural language processing. Nevertheless, recent studies, e.g., Zhao et al. (2017) show that these techniques inadvertently capture the societal bias hidden in the corpus and further amplify it. However, their analysis is conducted only on models' top predictions. In this paper, we investigate the gender bias amplification issue from the distribution perspective and demonstrate that the bias is amplified in the view of predicted probability distribution over labels. We further propose a bias mitigation approach based on post…
Pre-trained language models encode undesirable social biases, which are further exacerbated in downstream use. To this end, we propose MABEL (a Method for Attenuating Gender Bias using Entailment Labels), an intermediate pre-training approach for mitigating gender bias in contextualized representations. Key to our approach is the use of a contrastive learning objective on counterfactually augmented, gender-balanced entailment pairs from natural language inference (NLI) datasets. We also introduce an alignment regularizer that pulls identical entailment pairs along opposite gender directions cl…
Causally Testing Gender Bias in LLMs: A Case Study on Occupational Bias
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Columbia University（哥伦比亚大学）；RWTH Aachen（亚琛工业大学）；University of Michigan（密歇根大学）；ETH Zürich（苏黎世联邦理工学院）；Max Planck Institute（马克斯·普朗克研究所）
Wisdom from Diversity: Bias Mitigation Through Hybrid Human-LLM Crowds
Machine Learning Group, Université Libre de Bruxelles（布鲁塞尔自由大学机器学习组）；AI Lab, Vrije Universiteit Brussel（布鲁塞尔自由大学人工智能实验室）；FARI, AI for the Common-Good Institute, ULB-VUB（ULB-VUB共同利益人工智能研究所）；Center for Human-Compatible AI, UC Berkeley（伯克利人类兼容人工智能中心）
Microsoft AI Development Acceleration Program（微软人工智能发展加速计划）；Carnegie Mellon University（卡内基梅隆大学）
Online texts -- across genres, registers, domains, and styles -- are riddled with human stereotypes, expressed in overt or subtle ways. Word embeddings, trained on these texts, perpetuate and amplify these stereotypes, and propagate biases to machine learning models that use word embeddings as features. In this work, we propose a method to debias word embeddings in multiclass settings such as race and religion, extending the work of (Bolukbasi et al., 2016) from the binary setting, such as binary gender. Next, we propose a novel methodology for the evaluation of multiclass debiasing. We demons…
Data Bias Mitigation under Coverage Constraints & The Price of Fairness
Khoury College of Computer Sciences, Northeastern University（东北大学库里计算机科学学院）；Cheriton School of Computer Science, University of Waterloo（滑铁卢大学切里顿计算机科学学院）
Prioritizing fairness is of central importance in artificial intelligence (AI) systems, especially for those societal applications, e.g., hiring systems should recommend applicants equally from different demographic groups, and risk assessment systems must eliminate racism in criminal justice. Existing efforts towards the ethical development of AI systems have leveraged data science to mitigate biases in the training set or introduced fairness principles into the training process. For a deployed AI system, however, it may not allow for retraining or tuning in practice. By contrast, we propose …
Johannes Kepler University Linz（约翰内斯·开普勒林茨大学）；Know-Center GmbH（知识中心有限公司）
Societal biases resonate in the retrieved contents of information retrieval (IR) systems, resulting in reinforcing existing stereotypes. Approaching this issue requires established measures of fairness in respect to the representation of various social groups in retrieval results, as well as methods to mitigate such biases, particularly in the light of the advances in deep ranking models. In this work, we first provide a novel framework to measure the fairness in the retrieved text contents of ranking models. Introducing a ranker-agnostic measurement, the framework also enables the disentangle…
Carnegie Mellon University（卡内基梅隆大学）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Petuum Inc.（Petuum公司）
The proliferation of healthcare data has brought the opportunities of applying data-driven approaches, such as machine learning methods, to assist diagnosis. Recently, many deep learning methods have been shown with impressive successes in predicting disease status with raw input data. However, the "black-box" nature of deep learning and the high-reliability requirement of biomedical applications have created new challenges regarding the existence of confounding factors. In this paper, with a brief argument that inappropriate handling of confounding factors will lead to models' sub-optimal per…
