Calibration for the (Computationally-Identifiable) Masses

Abstract

As algorithms increasingly inform and influence decisions made about individuals, it becomes increasingly important to address concerns that these algorithms might be discriminatory. The output of an algorithm can be discriminatory for many reasons, most notably: (1) the data used to train the algorithm might be biased (in various ways) to favor certain populations over others; (2) the analysis of this training data might inadvertently or maliciously introduce biases that are not borne out in the data. This work focuses on the latter concern. We develop and study multicalbration -- a new measure of algorithmic fairness that aims to mitigate concerns about discrimination that is introduced in the process of learning a predictor from data. Multicalibration guarantees accurate (calibrated) predictions for every subpopulation that can be identified within a specified class of computations. We think of the class as being quite rich; in particular, it can contain many overlapping subgroups of a protected group. We show that in many settings this strong notion of protection from discrimination is both attainable and aligned with the goal of obtaining accurate predictions. Along the way, we present new algorithms for learning a multicalibrated predictor, study the computational complexity of this task, and draw new connections to computational learning models such as agnostic learning.

References in corpus (6)

• Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
• On Fairness and Calibration
• A Convex Framework for Fair Regression
• Fairness Through Computationally-Bounded Awareness
• Fairer and more accurate, but for whom?
• Lower Bound for Randomized First Order Convex Optimization

Cited by in corpus (22)

• Fairness in Machine Learning: A Survey
• Fairness in Machine Learning
• An Empirical Characterization of Fair Machine Learning For Clinical Risk Prediction
• Uncertainty Toolbox: an Open-Source Library for Assessing, Visualizing, and Improving Uncertainty Quantification
• Putting Fairness Principles into Practice: Challenges, Metrics, and Improvements
• Prediction uncertainty validation for computational chemists
• Review of Mathematical frameworks for Fairness in Machine Learning
• Beyond Pinball Loss: Quantile Methods for Calibrated Uncertainty Quantification
• SenSeI: Sensitive Set Invariance for Enforcing Individual Fairness
• Envy-Free Classification
• Counterfactual Reasoning for Fair Clinical Risk Prediction
• A survey of bias in Machine Learning through the prism of Statistical Parity for the Adult Data Set
• On the Role of Dataset Quality and Heterogeneity in Model Confidence
• Black Loans Matter: Distributionally Robust Fairness for Fighting Subgroup Discrimination
• Maximum Weighted Loss Discrepancy
• From Soft Classifiers to Hard Decisions: How fair can we be?
• Multi-Differential Fairness Auditor for Black Box Classifiers
• Stronger Calibration Lower Bounds via Sidestepping
• Algorithm Fairness in AI for Medicine and Healthcare
• Individually Fair Gradient Boosting
• Right Decisions from Wrong Predictions: A Mechanism Design Alternative to Individual Calibration
• Consider the Alternatives: Navigating Fairness-Accuracy Tradeoffs via Disqualification