Linear Classifiers that Encourage Constructive Adaptation
arXiv:2011.00355
Abstract
Machine learning systems are often used in settings where individuals adapt their features to obtain a desired outcome. In such settings, strategic behavior leads to a sharp loss in model performance in deployment. In this work, we aim to address this problem by learning classifiers that encourage decision subjects to change their features in a way that leads to improvement in both predicted \emph{and} true outcome. We frame the dynamics of prediction and adaptation as a two-stage game, and characterize optimal strategies for the model designer and its decision subjects. In benchmarks on simulated and real-world datasets, we find that classifiers trained using our method maintain the accuracy of existing approaches while inducing higher levels of improvement and less manipulation.
References in corpus (13)
- Equality of Opportunity in Supervised Learning
- Counterfactual Fairness
- Actionable Recourse in Linear Classification
- AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Bias
- The Disparate Effects of Strategic Manipulation
- Algorithmic recourse under imperfect causal knowledge: a probabilistic approach
- A survey of algorithmic recourse: definitions, formulations, solutions, and prospects
- Recommendations and User Agency: The Reachability of Collaboratively-Filtered Information
- Optimal Decision Making Under Strategic Behavior
- Repairing without Retraining: Avoiding Disparate Impact with Counterfactual Distributions
- Outside the Echo Chamber: Optimizing the Performative Risk
- Equalizing Recourse across Groups
- How to Learn when Data Reacts to Your Model: Performative Gradient Descent