Performative Prediction

Performative prediction studies how predictive models, when deployed, influence the very data they aim to predict, leading to feedback loops and distribution shifts. Current research focuses on developing algorithms and models, including those based on repeated retraining, stochastic gradient descent, and even reinforcement learning, that can find performatively stable solutions—models that are optimal given the data distribution they themselves induce—while addressing issues like fairness, bias amplification, and constraints. This field is significant because it tackles the limitations of traditional machine learning in dynamic, interactive settings, impacting areas such as algorithmic decision-making, online platforms, and policy design by improving model robustness and societal impact.

Papers

November 13, 2024

Microfoundation Inference for Strategic Prediction
Daniele Bracale, Subha Maity, Felipe Maia Polo, Seamus Somerstep, Moulinath Banerjee, Yuekai Sun
Human Prediction Predictive Model Prediction Task Performative Prediction Strategy Discovery

October 23, 2024

ProFL: Performative Robust Optimal Federated Learning
Xue Zheng, Tian Xie, Xuwei Tan, Aylin Yener, Xueru Zhang, Ali Payani, Myungjin Lee
Federated Learning Performative Prediction Robust Federated Learning Data Distribution Shift

August 9, 2024

Performative Prediction on Games and Mechanism Design
António Góis, Mehrnaz Mofakhami, Fernando P. Santos, Gauthier Gidel, Simon Lacoste-Julien
Video Game Game Theoretic Collective Behavior Mechanism Design Multi Agent Scenario Performative Prediction

June 24, 2024

Addressing Polarization and Unfairness in Performative Prediction
Kun Jin, Tian Xie, Yang Liu, Xueru Zhang
Political Polarization Performative Prediction Fairness Intervention Fairness Method Group Loss Data Distribution Shift

May 24, 2024

Learning the Distribution Map in Reverse Causal Performative Prediction
Daniele Bracale, Subha Maity, Moulinath Banerjee, Yuekai Sun
Distribution Shift Causal Model Distributional Assumption Performative Prediction Macroeconomic Model Distribution Shift Problem

May 23, 2024

Understanding the Training and Generalization of Pretrained Transformer for Sequential Decision Making
Hanzhao Wang, Yu Pan, Fupeng Sun, Shang Liu, Kalyan Talluri, Guanting Chen, Xiaocheng Li
Reinforcement Learning Strong Generalization Training Data Sequential Decision Making Problem Sequential Decision Making Optimal Action Performative Prediction

April 17, 2024

Clipped SGD Algorithms for Privacy Preserving Performative Prediction: Bias Amplification and Remedies
Qiang Li, Michal Yemini, Hoi-To Wai
Optimization Purpose Stochastic Gradient Descent Performative Prediction Clipped Stochastic Gradient Descent Bias Amplification Fast AltGDA Type Algorithm Stable Optimization

March 1, 2024

Evaluating and Correcting Performative Effects of Decision Support Systems via Causal Domain Shift
Philip Boeken, Onno Zoeter, Joris M. Mooij
Human Prediction Decision Support System Performative Prediction Crime Prediction

February 15, 2024

Performative Reinforcement Learning in Gradually Shifting Environments
Ben Rank, Stelios Triantafyllou, Debmalya Mandal, Goran Radanovic
Reinforcement Learning Continual Training Performative Prediction Time Varying Environment

October 25, 2023

Performative Prediction: Past and Future
Moritz Hardt, Celestine Mendler-Dünner
Human Prediction Future Reasoning Performative Prediction Performative Power False Negative Prediction

October 9, 2023

Performative Time-Series Forecasting
Zhiyuan Zhao, Alexander Rodriguez, B. Aditya Prakash
Time Series Forecasting Performative Prediction Performative Power

September 22, 2023

Zero-Regret Performative Prediction Under Inequality Constraints
Wenjing Yan, Xuanyu Cao
Primal Dual Algorithm Performative Prediction Low Regret Inequality Constraint Stochastic Primal Dual

May 30, 2023

Plug-in Performative Optimization
Licong Lin, Tijana Zrnic
Performative Prediction Best Response Explicit Feedback Based Recommender System

May 28, 2023

Incentivizing honest performative predictions with proper scoring rules
Caspar Oesterheld, Johannes Treutlein, Emery Cooper, Rubi Hudson
Human Prediction Near Optimality Performative Prediction Scoring Rule Optimal Data

May 8, 2023

Performative Federated Learning: A Solution to Model-Dependent and Heterogeneous Distribution Shifts
Kun Jin, Tongxin Yin, Zhongzhu Chen, Zeyu Sun, Xueru Zhang, Yang Liu, Mingyan Liu
Federated Learning Distribution Shift Solution Path FedAvg Algorithm Performative Prediction Data Distribution Shift Stable Optimization

May 1, 2023

Performative Prediction with Bandit Feedback: Learning through Reparameterization
Yatong Chen, Wei Tang, Chien-Ju Ho, Yang Liu
Bandit Feedback Regret Guarantee Reparameterization Method Performative Prediction Social Prediction

April 14, 2023

Performative Prediction with Neural Networks
Mehrnaz Mofakhami, Ioannis Mitliagkas, Gauthier Gidel
Neural Network Loss Function Distribution Shift Performative Prediction Stable Classifier

October 4, 2022

Making Decisions under Outcome Performativity
Michael P. Kim, Juan C. Perdomo
Decision Relevant Information Performative Prediction Performative Power

September 2, 2022

Optimizing the Performative Risk under Weak Convexity Assumptions
Yulai Zhao
Supervised Learning Performative Prediction Strong Convexity Convex Loss Function Minimization Problem

July 12, 2022

Long Term Fairness for Minority Groups via Performative Distributionally Robust Optimization
Liam Peet-Pare, Nidhi Hegde, Alona Fyshe
Machine Learning Distributionally Robust Optimization Performative Prediction Fairness Research Long Term Fairness Fairness Criterion Minority Group