Timezone: »

Performative Prediction
Juan Perdomo · Tijana Zrnic · Celestine Mendler-Dünner · Moritz Hardt

Thu Jul 16 06:00 AM -- 06:45 AM & Thu Jul 16 07:00 PM -- 07:45 PM (PDT) @

When predictions support decisions they may influence the outcome they aim to predict. We call such predictions performative; the prediction influences the target. Performativity is a well-studied phenomenon in policy-making that has so far been neglected in supervised learning. When ignored, performativity surfaces as undesirable distribution shift, routinely addressed with retraining. We develop a risk minimization framework for performative prediction bringing together concepts from statistics, game theory, and causality. A conceptual novelty is an equilibrium notion we call performative stability. Performative stability implies that the predictions are calibrated not against past outcomes, but against the future outcomes that manifest from acting on the prediction. Our main results are necessary and sufficient conditions for the convergence of retraining to a performatively stable point of nearly minimal loss. In full generality, performative prediction strictly subsumes the setting known as strategic classification. We thus also give the first sufficient conditions for retraining to overcome strategic feedback effects.

Author Information

Juan Perdomo (University of California, Berkeley)
Tijana Zrnic (University of California, Berkeley)
Celestine Mendler-Dünner (University of California, Berkeley)
Moritz Hardt (University of California, Berkeley)

More from the Same Authors