Do the Ends Justify the Means? Variation in the Distributive and Procedural Fairness of Machine Learning Algorithms
Abstract Recent advances in machine learning methods have created opportunities to eliminate unfairness from algorithmic decision making. Multiple computational techniques (i.e., algorithmic fairness criteria) have arisen out of this work. Yet, urgent questions remain about the percei...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer Netherlands
2022
|
| Online Access: | https://hdl.handle.net/1721.1/146834 |
| _version_ | 1811078575894298624 |
|---|---|
| author | Morse, Lily Teodorescu, Mike H. M. Awwad, Yazeed Kane, Gerald C. |
| author2 | Massachusetts Institute of Technology. Media Laboratory |
| author_facet | Massachusetts Institute of Technology. Media Laboratory Morse, Lily Teodorescu, Mike H. M. Awwad, Yazeed Kane, Gerald C. |
| author_sort | Morse, Lily |
| collection | MIT |
| description | Abstract
Recent advances in machine learning methods have created opportunities to eliminate unfairness from algorithmic decision making. Multiple computational techniques (i.e., algorithmic fairness criteria) have arisen out of this work. Yet, urgent questions remain about the perceived fairness of these criteria and in which situations organizations should use them. In this paper, we seek to gain insight into these questions by exploring fairness perceptions of five algorithmic criteria. We focus on two key dimensions of fairness evaluations: distributive fairness and procedural fairness. We shed light on variation in the potential for different algorithmic criteria to facilitate distributive fairness. Subsequently, we discuss procedural fairness and provide a framework for understanding how algorithmic criteria relate to essential aspects of this construct, which helps to identify when a specific criterion is suitable. From a practical standpoint, we encourage organizations to recognize that managing fairness in machine learning systems is complex, and that adopting a blind or one-size-fits-all mentality toward algorithmic criteria will surely damage people’s attitudes and trust in automated technology. Instead, firms should carefully consider the subtle yet significant differences between these technical solutions. |
| first_indexed | 2024-09-23T11:02:27Z |
| format | Article |
| id | mit-1721.1/146834 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T11:02:27Z |
| publishDate | 2022 |
| publisher | Springer Netherlands |
| record_format | dspace |
| spelling | mit-1721.1/1468342023-07-05T20:34:11Z Do the Ends Justify the Means? Variation in the Distributive and Procedural Fairness of Machine Learning Algorithms Morse, Lily Teodorescu, Mike H. M. Awwad, Yazeed Kane, Gerald C. Massachusetts Institute of Technology. Media Laboratory Abstract Recent advances in machine learning methods have created opportunities to eliminate unfairness from algorithmic decision making. Multiple computational techniques (i.e., algorithmic fairness criteria) have arisen out of this work. Yet, urgent questions remain about the perceived fairness of these criteria and in which situations organizations should use them. In this paper, we seek to gain insight into these questions by exploring fairness perceptions of five algorithmic criteria. We focus on two key dimensions of fairness evaluations: distributive fairness and procedural fairness. We shed light on variation in the potential for different algorithmic criteria to facilitate distributive fairness. Subsequently, we discuss procedural fairness and provide a framework for understanding how algorithmic criteria relate to essential aspects of this construct, which helps to identify when a specific criterion is suitable. From a practical standpoint, we encourage organizations to recognize that managing fairness in machine learning systems is complex, and that adopting a blind or one-size-fits-all mentality toward algorithmic criteria will surely damage people’s attitudes and trust in automated technology. Instead, firms should carefully consider the subtle yet significant differences between these technical solutions. 2022-12-12T13:55:27Z 2022-12-12T13:55:27Z 2021-10-18 2022-12-10T04:21:48Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/146834 Morse, Lily, Teodorescu, Mike H. M., Awwad, Yazeed and Kane, Gerald C. 2021. "Do the Ends Justify the Means? Variation in the Distributive and Procedural Fairness of Machine Learning Algorithms." en https://doi.org/10.1007/s10551-021-04939-5 Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. The Author(s), under exclusive licence to Springer Nature B.V. application/pdf Springer Netherlands Springer Netherlands |
| spellingShingle | Morse, Lily Teodorescu, Mike H. M. Awwad, Yazeed Kane, Gerald C. Do the Ends Justify the Means? Variation in the Distributive and Procedural Fairness of Machine Learning Algorithms |
| title | Do the Ends Justify the Means? Variation in the Distributive and Procedural Fairness of Machine Learning Algorithms |
| title_full | Do the Ends Justify the Means? Variation in the Distributive and Procedural Fairness of Machine Learning Algorithms |
| title_fullStr | Do the Ends Justify the Means? Variation in the Distributive and Procedural Fairness of Machine Learning Algorithms |
| title_full_unstemmed | Do the Ends Justify the Means? Variation in the Distributive and Procedural Fairness of Machine Learning Algorithms |
| title_short | Do the Ends Justify the Means? Variation in the Distributive and Procedural Fairness of Machine Learning Algorithms |
| title_sort | do the ends justify the means variation in the distributive and procedural fairness of machine learning algorithms |
| url | https://hdl.handle.net/1721.1/146834 |
| work_keys_str_mv | AT morselily dotheendsjustifythemeansvariationinthedistributiveandproceduralfairnessofmachinelearningalgorithms AT teodorescumikehm dotheendsjustifythemeansvariationinthedistributiveandproceduralfairnessofmachinelearningalgorithms AT awwadyazeed dotheendsjustifythemeansvariationinthedistributiveandproceduralfairnessofmachinelearningalgorithms AT kanegeraldc dotheendsjustifythemeansvariationinthedistributiveandproceduralfairnessofmachinelearningalgorithms |