Measuring the Biases that Matter: The Ethical and Causal Foundations for Measures of Fairness in Algorithms
Proceedings of the Conference on Fairness, Accountability, and Transparency 2019:269-278 (2019)
AbstractMeasures of algorithmic bias can be roughly classified into four categories, distinguished by the conditional probabilistic dependencies to which they are sensitive. First, measures of "procedural bias" diagnose bias when the score returned by an algorithm is probabilistically dependent on a sensitive class variable (e.g. race or sex). Second, measures of "outcome bias" capture probabilistic dependence between class variables and the outcome for each subject (e.g. parole granted or loan denied). Third, measures of "behavior-relative error bias" capture probabilistic dependence between class variables and the algorithmic score, conditional on target behaviors (e.g. recidivism or loan default). Fourth, measures of "score-relative error bias" capture probabilistic dependence between class variables and behavior, conditional on score. Several recent discussions have demonstrated a tradeoff between these different measures of algorithmic bias, and at least one recent paper has suggested conditions under which tradeoffs may be minimized. In this paper we use the machinery of causal graphical models to show that, under standard assumptions, the underlying causal relations among variables forces some tradeoffs. We delineate a number of normative considerations that are encoded in different measures of bias, with reference to the philosophical literature on the wrongfulness of disparate treatment and disparate impact. While both kinds of error bias are nominally motivated by concern to avoid disparate impact, we argue that consideration of causal structures shows that these measures are better understood as complicated and unreliable measures of procedural biases (i.e. disparate treatment). Moreover, while procedural bias is indicative of disparate treatment, we show that the measure of procedural bias one ought to adopt is dependent on the account of the wrongfulness of disparate treatment one endorses. Finally, given that neither score-relative nor behavior-relative measures of error bias capture the relevant normative considerations, we suggest that error bias proper is best measured by score-based measures of accuracy, such as the Brier score.
Added to PP
Historical graph of downloads
References found in this work
No references found.
Citations of this work
Ontology, Neural Networks, and the Social Sciences.David Strohmaier - 2020 - Synthese 199 (1-2):4775-4794.
Similar books and articles
Measuring Fairness in an Unfair World.Jonathan Herington - 2020 - Proceedings of AAAI/ACM Conference on AI, Ethics, and Society 2020:286-292.
Algorithmic Fairness in Mortgage Lending: From Absolute Conditions to Relational Trade-Offs.Michelle Seng Ah Lee & Luciano Floridi - 2021 - Minds and Machines 31 (1):165-191.
Fair, Transparent, and Accountable Algorithmic Decision-Making Processes: The Premise, the Proposed Solutions, and the Open Challenges.Bruno Lepri, Nuria Oliver, Emmanuel Letouzé, Alex Pentland & Patrick Vinck - 2018 - Philosophy and Technology 31 (4):611-627.
Formalising Trade-Offs Beyond Algorithmic Fairness: Lessons From Ethical Philosophy and Welfare Economics.Michelle Seng Ah Lee, Luciano Floridi & Jatinder Singh - 2021 - AI and Ethics 3.
Understanding Perception of Algorithmic Decisions: Fairness, Trust, and Emotion in Response to Algorithmic Management.Min Kyung Lee - 2018 - Big Data and Society 5 (1).
A Moral Framework for Understanding of Fair ML Through Economic Models of Equality of Opportunity.Hoda Heidari - 2019 - Proceedings of the Conference on Fairness, Accountability, and Transparency 1.
What's Fair About Individual Fairness?Will Fleisher - 2021 - Proceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society.
Fairness, Feelings, and Ethical Decision- Making: Consequences of Violating Community Standards of Fairness.Maurice E. Schweitzer & Donald E. Gibson - 2008 - Journal of Business Ethics 77 (3):287-301.
On Statistical Criteria of Algorithmic Fairness.Brian Hedden - 2021 - Philosophy and Public Affairs 49 (2):209-231.
Algorithmic Bias and the Value Sensitive Design Approach.Judith Simon, Pak-Hang Wong & Gernot Rieder - 2020 - Internet Policy Review 9 (4).
Fairness, Feelings, and Ethical Decision- Making: Consequences of Violating Community Standards of Fairness. [REVIEW]Maurice E. Schweitzer & Donald E. Gibson - 2008 - Journal of Business Ethics 77 (3):287 - 301.
Fairer Machine Learning in the Real World: Mitigating Discrimination Without Collecting Sensitive Data.Reuben Binns & Michael Veale - 2017 - Big Data and Society 4 (2).
Procedural and Distributive Fairness: Determinants of Overall Price Fairness.Jodie L. Ferguson, Pam Scholder Ellen & William O. Bearden - 2014 - Journal of Business Ethics 121 (2):1-15.
Algorithms and Autonomy: The Ethics of Automated Decision Systems.Alan Rubel, Clinton Castro & Adam Pham - 2021 - Cambridge University Press.