Alternative Microfoundations for Strategic Classification

Meena Jagadeesan, Celestine Mendler-Dünner, Moritz Hardt
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:4687-4697, 2021.

Abstract

When reasoning about strategic behavior in a machine learning context it is tempting to combine standard microfoundations of rational agents with the statistical decision theory underlying classification. In this work, we argue that a direct combination of these ingredients leads to brittle solution concepts of limited descriptive and prescriptive value. First, we show that rational agents with perfect information produce discontinuities in the aggregate response to a decision rule that we often do not observe empirically. Second, when any positive fraction of agents is not perfectly strategic, desirable stable points—where the classifier is optimal for the data it entails—no longer exist. Third, optimal decision rules under standard microfoundations maximize a measure of negative externality known as social burden within a broad class of assumptions about agent behavior. Recognizing these limitations we explore alternatives to standard microfoundations for binary classification. We describe desiderata that help navigate the space of possible assumptions about agent responses, and we then propose the noisy response model. Inspired by smoothed analysis and empirical observations, noisy response incorporates imperfection in the agent responses, which we show mitigates the limitations of standard microfoundations. Our model retains analytical tractability, leads to more robust insights about stable points, and imposes a lower social burden at optimality.

Cite this Paper


BibTeX
@InProceedings{pmlr-v139-jagadeesan21a, title = {Alternative Microfoundations for Strategic Classification}, author = {Jagadeesan, Meena and Mendler-D{\"u}nner, Celestine and Hardt, Moritz}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {4687--4697}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/jagadeesan21a/jagadeesan21a.pdf}, url = {https://proceedings.mlr.press/v139/jagadeesan21a.html}, abstract = {When reasoning about strategic behavior in a machine learning context it is tempting to combine standard microfoundations of rational agents with the statistical decision theory underlying classification. In this work, we argue that a direct combination of these ingredients leads to brittle solution concepts of limited descriptive and prescriptive value. First, we show that rational agents with perfect information produce discontinuities in the aggregate response to a decision rule that we often do not observe empirically. Second, when any positive fraction of agents is not perfectly strategic, desirable stable points—where the classifier is optimal for the data it entails—no longer exist. Third, optimal decision rules under standard microfoundations maximize a measure of negative externality known as social burden within a broad class of assumptions about agent behavior. Recognizing these limitations we explore alternatives to standard microfoundations for binary classification. We describe desiderata that help navigate the space of possible assumptions about agent responses, and we then propose the noisy response model. Inspired by smoothed analysis and empirical observations, noisy response incorporates imperfection in the agent responses, which we show mitigates the limitations of standard microfoundations. Our model retains analytical tractability, leads to more robust insights about stable points, and imposes a lower social burden at optimality.} }
Endnote
%0 Conference Paper %T Alternative Microfoundations for Strategic Classification %A Meena Jagadeesan %A Celestine Mendler-Dünner %A Moritz Hardt %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-jagadeesan21a %I PMLR %P 4687--4697 %U https://proceedings.mlr.press/v139/jagadeesan21a.html %V 139 %X When reasoning about strategic behavior in a machine learning context it is tempting to combine standard microfoundations of rational agents with the statistical decision theory underlying classification. In this work, we argue that a direct combination of these ingredients leads to brittle solution concepts of limited descriptive and prescriptive value. First, we show that rational agents with perfect information produce discontinuities in the aggregate response to a decision rule that we often do not observe empirically. Second, when any positive fraction of agents is not perfectly strategic, desirable stable points—where the classifier is optimal for the data it entails—no longer exist. Third, optimal decision rules under standard microfoundations maximize a measure of negative externality known as social burden within a broad class of assumptions about agent behavior. Recognizing these limitations we explore alternatives to standard microfoundations for binary classification. We describe desiderata that help navigate the space of possible assumptions about agent responses, and we then propose the noisy response model. Inspired by smoothed analysis and empirical observations, noisy response incorporates imperfection in the agent responses, which we show mitigates the limitations of standard microfoundations. Our model retains analytical tractability, leads to more robust insights about stable points, and imposes a lower social burden at optimality.
APA
Jagadeesan, M., Mendler-Dünner, C. & Hardt, M.. (2021). Alternative Microfoundations for Strategic Classification. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:4687-4697 Available from https://proceedings.mlr.press/v139/jagadeesan21a.html.

Related Material