Aequitas: A Bias and Fairness Audit Toolkit

Saleiro, Pedro; Kuester, Benedict; Loren, Hinkson,; Jesse, London,; Stevens, Abby; Anisfeld, Ari; Rodolfa, Kit T.; Ghani, Rayid

doi:10.48550/arxiv.1811.05577

Cited by 78 publications

(89 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…On the other hand, TT3 applied mean-imputation and TT4 applied both median-and mode-imputation using df.fillna(), which exhibits fairness compared to data removal. While our findings suggest that removing data items with MV introduces bias, the most popular fairness tools AIF 360 [8], Aequitas [58], Themis-ML [6] ignore these data items and remove entire row/column. Our evaluation strategy confirms that the tools can integrate existing imputation methods [62] in the pipeline and allow users to choose appropriate ones.…”

Section: Fairness Analysis Of Stagesmentioning

confidence: 66%

Fair preprocessing: towards understanding compositional fairness of data transformers in machine learning pipeline

Biswas

Rajan

2021

Proceedings of the 29th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Softw

View full text Add to dashboard Cite

In recent years, many incidents have been reported where machine learning models exhibited discrimination among people based on race, sex, age, etc. Research has been conducted to measure and mitigate unfairness in machine learning models. For a machine learning task, it is a common practice to build a pipeline that includes an ordered set of data preprocessing stages followed by a classifier. However, most of the research on fairness has considered a single classifier based prediction task. What are the fairness impacts of the preprocessing stages in machine learning pipeline? Furthermore, studies showed that often the root cause of unfairness is ingrained in the data itself, rather than the model. But no research has been conducted to measure the unfairness caused by a specific transformation made in the data preprocessing stage. In this paper, we introduced the causal method of fairness to reason about the fairness impact of data preprocessing stages in ML pipeline. We leveraged existing metrics to define the fairness measures of the stages. Then we conducted a detailed fairness evaluation of the preprocessing stages in 37 pipelines collected from three different sources. Our results show that certain data transformers are causing the model to exhibit unfairness. We identified a number of fairness patterns in several categories of data transformers. Finally, we showed how the local fairness of a preprocessing stage composes in the global fairness of the pipeline. We used the fairness composition to choose appropriate downstream transformer that mitigates unfairness in the machine learning pipeline. CCS CONCEPTS• Software and its engineering → Software creation and management; • Computing methodologies → Machine learning.

show abstract

Section: Fairness Analysis Of Stagesmentioning

confidence: 66%

Fair preprocessing: towards understanding compositional fairness of data transformers in machine learning pipeline

Biswas

Rajan

2021

Proceedings of the 29th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Softw

View full text Add to dashboard Cite

show abstract

“…In this sub-section, we present some of the related work that deals with maintaining as well as investigating the fairness of data in the AI domain and are listed in [20][21][22][23][24][25][26][27].…”

Section: Related Workmentioning

confidence: 99%

“…The framework is tested on the problem of fairly predicting the acceptance of law students. In addition, in [26,27], an AI-based real-time toolkit for fairness assurance is presented.…”

Section: Related Workmentioning

confidence: 99%

“…In [26], the Aequitas toolkit is presented, which allows users to test models with respect to various bias and fairness metrics for different population subgroups. In [27], the AI Fairness 360 (AIF360) toolkit developed by IBM is presented.…”

Section: Related Workmentioning

confidence: 99%

“…The work presented in [20][21][22][23][24][25][26][27] addressed fairness with respect to various demographic groups, which cannot be used to ensure fairness in wireless data because they do not contain demographic information. Inspired by the group fairness shown in [20][21][22][23][24][25][26][27], this work proposes a wireless dataset D categorization and combination scheme to ensure fairness in selecting UEs for packet transmission at any scheduling epochs.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

A Fairness of Data Combination in Wireless Packet Scheduling

Bhandari

Ranjan

Kim

et al. 2022

Sensors

View full text Add to dashboard Cite

With the proliferation of artificial intelligence (AI) technology, the function of AI in a sixth generation (6G) environment is likely to come into play on a large scale. Moreover, in recent years, with the rapid advancement in AI technology, the ethical issues of AI have become a hot topic. In this paper, the ethical concern of AI in wireless networks is studied from the perspective of fairness in data. To make the dataset fairer, novel dataset categorization and dataset combination schemes are proposed. For the dataset categorization scheme, a deep-learning-based dataset categorization (DLDC) model is proposed. Based on the results of the DLDC model, the input dataset is categorized based on the group index. The datasets based on the group index are combined using various combination schemes. Through simulations, the results of each dataset combination method and their performance are compared, and the advantages and disadvantages of fairness and performance according to the dataset configuration are analyzed.

show abstract

Opportunistic screening for low bone density using abdominopelvic computed tomography scans

et al. 2023

View full text Add to dashboard Cite

Background While low bone density is a major burden on US health system, current osteoporosis screening guidelines by the US Preventive Services Task Force are limited to women aged ≥65 and all postmenopausal women with certain risk factors. Even within recommended screening groups, actual screening rates are low (<26%) and vary across socioeconomic groups. The proposed model can opportunistically screen patients using abdominal CT studies for low bone density who may otherwise go undiagnosed. Purpose To develop an artificial intelligence (AI) model for opportunistic screening of low bone density using both contrast and non‐contrast abdominopelvic computed tomography (CT) exams, for the purpose of referral to traditional bone health management, which typically begins with dual energy X‐ray absorptiometry (DXA). Methods We collected 6083 contrast‐enhanced CT imaging exams paired with DXA exams within ±6 months documented between May 2015 and August 2021 in a single institution with four major healthcare practice regions. Our fusion AI pipeline receives the coronal and axial plane images of a contrast enhanced abdominopelvic CT exam and basic patient demographics (age, gender, body cross section lengths) to predict risk of low bone mass. The models were trained on lumbar spine T‐scores from DXA exams and tested on multi‐site imaging exams. The model was again tested in a prospective group (N = 344) contrast‐enhanced and non‐contrast‐enhanced studies. Results The models were evaluated on the same test set (1208 exams)—(1) Baseline model using demographic factors from electronic medical records (EMR) ‐ 0.7 area under the curve of receiver operator characteristic (AUROC); Imaging based models: (2) axial view ‐ 0.83 AUROC; (3) coronal view‐ 0.83 AUROC; (4) Fusion model—Imaging + demographic factors ‐ 0.86 AUROC. The prospective test yielded one missed positive DXA case with a hip prosthesis among 23 positive contrast‐enhanced CT exams and 0% false positive rate for non‐contrast studies. Both positive cases among non‐contrast enhanced CT exams were successfully detected. While only about 8% patients from prospective study received a DXA exam within 2 years, about 30% were detected with low bone mass by the fusion model, highlighting the need for opportunistic screening. Conclusions The fusion model, which combines two planes of CT images and EMRs data, outperformed individual models and provided a high, robust diagnostic performance for opportunistic screening of low bone density using contrast and non‐contrast CT exams. This model could potentially improve bone health risk assessment with no additional cost. The model's handling of metal implants is an ongoing effort.

show abstract

Aequitas: A Bias and Fairness Audit Toolkit

Cited by 78 publications

References 10 publications

Fair preprocessing: towards understanding compositional fairness of data transformers in machine learning pipeline

Fair preprocessing: towards understanding compositional fairness of data transformers in machine learning pipeline

A Fairness of Data Combination in Wireless Packet Scheduling

Opportunistic screening for low bone density using abdominopelvic computed tomography scans

Contact Info

Product

Resources

About