Rethinking Fairness: An Interdisciplinary Survey of Critiques of Hegemonic ML Fairness Approaches

Authors: Lindsay Weinberg

JAIR 2022 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical This survey article assesses and compares existing critiques of current fairness-enhancing technical interventions in machine learning (ML) that draw from a range of non-computing disciplines, including philosophy, feminist studies, critical race and ethnic studies, legal studies, anthropology, and science and technology studies.
Researcher Affiliation Academia Lindsay Weinberg EMAIL Honors College, Purdue University West Lafayette, IN 47906
Pseudocode No The paper is a survey and does not propose or describe any specific algorithms or methods in pseudocode format. It analyzes existing critiques and proposed solutions conceptually.
Open Source Code No This article is a survey of existing critiques of ML fairness approaches and does not present a new methodology that would require open-source code. It refers to open-source toolkits by third-parties, such as 'IBM s AI Fairness 360 for mitigating bias in training data (Varshney, 2018) and Google s ML-fairness-gym for anticipating the effects of automated decision systems (Srinivasan, 2020)', but does not provide code for its own content.
Open Datasets No This article is a survey and does not conduct original experiments using specific datasets. It discusses datasets and data collection practices in other research, but does not provide access information for a dataset used in its own analysis.
Dataset Splits No As a survey article, this paper does not involve experimental data processing or machine learning model training, and therefore does not specify dataset splits.
Hardware Specification No This paper is a survey article and does not describe experimental work requiring specific hardware. Therefore, no hardware specifications are provided.
Software Dependencies No As a survey article, this paper does not involve the implementation of new software or methodologies that would require specific software dependencies and version numbers.
Experiment Setup No This paper is a survey and does not describe experimental work requiring specific experimental setup details, hyperparameters, or training configurations.