Created by Murat Durmus (CEO AISOMA)

LinkedIn: https://www.linkedin.com/in/ceosaisoma/

An overview of some useful

Fairness Frameworks & Packages

Content

1. The LinkedIn Fairness Toolkit (LiFT) ................................................................................... 1

2. Fairness-indicators: Tensorflow's Fairness Evaluation and Visualization Toolkit
(Google) ............................................................................................................................................ 2
3. AI Fairness 360 (IBM) ............................................................................................................... 3
4. Fairlearn: Fairness in machine learning mitigation algorithms (Microsoft) ............. 3
5. Algofairness ................................................................................................................................ 3
6. FairSight: Visual Analytics for Fairness in Decision Making ........................................ 4
7. Aequitas: Bias and Fairness Audit Toolkit ......................................................................... 5
8. CERTIFAI: A Common Framework to Provide Explanations and Analyse the
Fairness and Robustness of Black-box Models .................................................................... 6
9. ML-fairness-gym: Google's implementation based on OpenAI's Gym ....................... 6
10. scikit-fairness........................................................................................................................... 7
11. Mitigating Gender Bias In Captioning System ................................................................ 7

1. The LinkedIn Fairness Toolkit (LiFT)

The LinkedIn Fairness Toolkit (LiFT) is a Scala/Spark library that enables the
measurement of fairness in large scale machine learning workflows. The library
can be deployed in training and scoring workflows to measure biases in training
data, evaluate fairness metrics for ML models, and detect statistically significant
1
 Created by Murat Durmus (CEO AISOMA)
LinkedIn: https://www.linkedin.com/in/ceosaisoma/

differences in their performance across different subgroups. It can also be used

for ad-hoc fairness analysis.

This library was created by Sriram Vasudevan and Krishnaram

Kenthapadi (work done while at LinkedIn).

More info: https://github.com/linkedin/LiFT

2. Fairness-indicators: Tensorflow's Fairness Evaluation and

Visualization Toolkit (Google)

Fairness Indicators is designed to support teams in evaluating, improving, and

comparing models for fairness concerns in partnership with the broader
Tensorflow toolkit.

The tool is currently actively used internally by many of our products. We would
love to partner with you to understand where Fairness Indicators is most useful,
and where added functionality would be valuable. Please reach out
at [email protected].

More info: https://github.com/tensorflow/fairness-indicators

2
 Created by Murat Durmus (CEO AISOMA)
LinkedIn: https://www.linkedin.com/in/ceosaisoma/

3. AI Fairness 360 (IBM)

The AI Fairness 360 toolkit is an extensible open-source library containg

techniques developed by the research community to help detect and mitigate
bias in machine learning models throughout the AI application lifecycle. AI
Fairness 360 package is available in both Python and R.

The AI Fairness 360 package includes

· a comprehensive set of metrics for datasets and models to test for biases,

· explanations for these metrics, and

· algorithms to mitigate bias in datasets and models. It is designed to translate

algorithmic research from the lab into the actual practice of domains as wide-
ranging as finance, human capital management, healthcare, and education. We
invite you to use it and improve it.

More info: https://github.com/Trusted-AI/AIF360

4. Fairlearn: Fairness in machine learning mitigation algorithms

(Microsoft)

Fairlearn is a Python package that empowers developers of artificial

intelligence (AI) systems to assess their system's fairness and mitigate any
observed unfairness issues. Fairlearn contains mitigation algorithms as well as a
Jupyter widget for model assessment. Besides the source code, this repository
also contains Jupyter notebooks with examples of Fairlearn usage.

More Info: https://github.com/fairlearn/fairlearn

5. Algofairness

BlackBoxAuditing

This repository contains a sample implementation of Gradient Feature Auditing

(GFA) meant to be generalizable to most datasets. For more information on the
repair process, see our paper on Certifying and Removing Disparate Impact. For
information on the full auditing process, see our paper on Auditing Black-box
Models for Indirect Influence.

3
 Created by Murat Durmus (CEO AISOMA)
LinkedIn: https://www.linkedin.com/in/ceosaisoma/

More info: https://github.com/algofairness/BlackBoxAuditing

fairness-comparison

This repository is meant to facilitate the benchmarking of fairness aware

machine learning algorithms.

The associated paper is:

A comparative study of fairness-enhancing interventions in machine learning by

Sorelle A. Friedler, Carlos Scheidegger, Suresh Venkatasubramanian, Sonam
Choudhary, Evan P. Hamilton, and Derek
Roth. https://arxiv.org/abs/1802.04422

More info: https://github.com/algofairness/fairness-comparison

fatconference-2019-toolkit-tutorial

More info: https://github.com/algofairness/fatconference-2019-

toolkit-tutorial

fatconference-2018-auditing-tutorial

More info: https://github.com/algofairness/fatconference-2018-

auditing-tutorial

runaway-feedback-loops-src

More info: https://github.com/algofairness/runaway-feedback-

loops-src

Knight

More info: https://github.com/algofairness/knight

6. FairSight: Visual Analytics for Fairness in Decision Making

FairSight is a viable fair decision-making system to assist decision makers in

achieving fair decision making through the machine learning workflow.

Specification

4
 Created by Murat Durmus (CEO AISOMA)
LinkedIn: https://www.linkedin.com/in/ceosaisoma/

React: Frontend framework for rendering and communicating with data

django: Python-based backend framework for serving API of data and running
machine learning work

scss: The stylesheet grammar for more flexible structure

d3.js: Javascript-based visualization library

More info: https://github.com/ayong8/FairSight

7. Aequitas: Bias and Fairness Audit Toolkit

Aequitas is an open-source bias audit toolkit for data scientists, machine learning
researchers, and policymakers to audit machine learning models for
discrimination and bias, and to make informed and equitable decisions around
developing and deploying predictive tools.

More info: https://github.com/dssg/aequitas

5
 Created by Murat Durmus (CEO AISOMA)
LinkedIn: https://www.linkedin.com/in/ceosaisoma/

8. CERTIFAI: A Common Framework to Provide Explanations and

Analyse the Fairness and Robustness of Black-box Models

Concerns within the machine learning community and external pressures from
regulators over the vulnerabilities of machine learning algorithms have spurred
on the fields of explainability, robustness, and fairness. Often, issues in
explainability, robustness, and fairness are confined to their specific sub-fields
and few tools exist for model developers to use to simultaneously build their
modeling pipelines in a transparent, accountable, and fair way. This can lead to
a bottleneck on the model developer’s side as they must juggle multiple methods
to evaluate their algorithms. In this paper, we present a single framework for
analyzing the robustness, fairness, and explainability of a classifier. The
framework, which is based on the generation of counterfactual explanations1
through a custom genetic algorithm, is flexible, model-agnostic, and does not
require access to model internals. The framework allows the user to calculate
robustness and fairness scores for individual models and generate explanations
for individual predictions which provide a means for actionable recourse
(changes to an input to help get a desired outcome). This is the first time that a
unified tool has been developed to address three key issues pertaining towards
building a responsible artificial intelligence system.

More info: CERTIFAI: A Common Framework to Provide Explanations and

Analyse the Fairness and Robustness of Black-box Models

9. ML-fairness-gym: Google's implementation based on OpenAI's

Gym

6
 Created by Murat Durmus (CEO AISOMA)
LinkedIn: https://www.linkedin.com/in/ceosaisoma/

ML-fairness-gym is a set of components for building simple simulations that

explore the potential long-run impacts of deploying machine learning-based
decision systems in social environments. As the importance of machine learning
fairness has become increasingly apparent, recent research has focused on
potentially surprising long term behaviors of enforcing measures of fairness that
were originally defined in a static setting. Key findings have shown that under
specific assumptions in simplified dynamic simulations, long term effects may in
fact counteract the desired goals.

More info: https://github.com/google/ml-fairness-gym

10. scikit-fairness

The goal of this project is to attempt to consolidate fairness related metrics,

transformers and models into a package that (hopefully) will become a
contribution project to scikit-learn.

Fairness, in data science, is a complex unsolved problem for which many tactics
are proposed - each with their own advantage and disadvantages. This packages
aims to make these tactics readily available, therefore enabling users to try and
evaluate different fairness techniques.

More info: https://github.com/koaning/scikit-fairness

11. Mitigating Gender Bias In Captioning System

This is the pytorch implemention for paper “Mitigating Gender Bias In Captioning
system”. Recent studies have shown that captioning datasets, such as the COCO
dataset, may contain severe social bias which could potentially lead to
unintentional discrimination in learning models. In this work, we specifically
focus on the gender bias problem.

Image Captioning Model with Guided Attention

We propose a novel Guided Attention Image Captioning model (GAIC) to

mitigate gender bias by self-supervising on model’s visual attention. GAIC has
two complementary streams to encourage the model to explore correct gender
features. The training pipeline can seamlessly add extra supervision to accelerate
the self-exploration process. Besides, GAIC is model-agnostic and can be easily
applied to various captioning models.
7
 Created by Murat Durmus (CEO AISOMA)
LinkedIn: https://www.linkedin.com/in/ceosaisoma/

More info:

https://github.com/CaptionGenderBias2020/Mitigating_Gender_Bias_In_Ca
ptioning_System

This might be also of interest: A collection of useful Slides & Quotes on AI-
Ethics and XAI

8
 Created by Murat Durmus (CEO AISOMA)
LinkedIn: https://www.linkedin.com/in/ceosaisoma/

aisoma.de

Contact

https://www.linkedin.com/in/ceosaisoma/

[email protected]

https://www.aisoma.de