More Like this

1. Stochastic Optimization for Performative Prediction

   Mendler-Dünner, Celestine ; Perdomo, Juan C. ; Zrnic, Tijana ; Hardt, Moritz ( July 2020 , Advances in Neural Information Processing Systems 33 (NeurIPS 2020))

   null (Ed.)

   In performative prediction, the choice of a model influences the distribution of future data, typically through actions taken based on the model's predictions. We initiate the study of stochastic optimization for performative prediction. What sets this setting apart from traditional stochastic optimization is the difference between merely updating model parameters and deploying the new model. The latter triggers a shift in the distribution that affects future data, while the former keeps the distribution as is. Assuming smoothness and strong convexity, we prove rates of convergence for both greedily deploying models after each stochastic update (greedy deploy) as well as for taking several updates before redeploying (lazy deploy). In both cases, our bounds smoothly recover the optimal O(1/k) rate as the strength of performativity decreases. Furthermore, they illustrate how depending on the strength of performative effects, there exists a regime where either approach outperforms the other. We experimentally explore the trade-off on both synthetic data and a strategic classification simulator. 

   more » « less
2. Multiplayer Performative Prediction: Learning in Decision-Dependent Games

   Narang, Adhyyan ; Faulkner, Evan ; Drusvyatskiy, Dmitriy ; Fazel, Maryam ; Ratliff, Lillian J. ( January 2023 , Journal of Machine Learning Research)

   Learning problems commonly exhibit an interesting feedback mechanism wherein the population data reacts to competing decision makers’ actions. This paper formulates a new game theoretic framework for this phenomenon, called multi-player performative prediction. We focus on two distinct solution concepts, namely (i) performatively stable equilibria and (ii) Nash equilibria of the game. The latter equilibria are arguably more informative, but are generally computationally difficult to find since they are solutions of nonmonotone games. We show that under mild assumptions, the performatively stable equilibria can be found efficiently by a variety of algorithms, including repeated retraining and the repeated (stochastic) gradient method. We then establish transparent sufficient conditions for strong monotonicity of the game and use them to develop algorithms for finding Nash equilibria. We investigate derivative free methods and adaptive gradient algorithms wherein each player alternates between learning a parametric description of their distribution and gradient steps on the empirical risk. Synthetic and semi-synthetic numerical experiments illustrate the results. 

   more » « less
3. How the map becomes the territory: prediction, performativity and the process of taking digital twins for granted

   https://doi.org/10.1007/s41469-024-00164-2  

   Leonardi, Paul M. ; Leavell, Virginia ( February 2024 , Journal of Organization Design)

   A growing body of literature argues that digital models do not just help organizational leaders to predict the future. Models can inadvertently produce the very future they purport to describe. In this view,performativityis a side-effect of digital modeling. But digital twins turn such thinking on its head. Digital twins are digital models that are designed to be performative—changes in the model are supposed to produce corresponding changes in the world the model represents. This is what makes digital twins useful. But for decision-makers to act in ways that align the world outside the model with the predictions contained within, they must first believe that the model is a faithful representation. In other words, for a digital twin to become performative, it must first be taken-for-granted as “real”. In this paper, we explore the technological and organizational characteristics that are likely to shape the level of taken-for-grantedness of a digital twin.

    

   more » « less
4. Strategic Behavior in Two-sided Matching Markets with Recommendation-enhanced Preference-formation

   Ionescu, Stefania ; Du, Yuhao ; Joseph, Kenneth ; Hannak, Aniko ( December 2023 , 37th Conference on Neural Information Processing Systems (NeurIPS 2023))

   Two-sided matching markets have long existed to pair agents in the absence of regulated exchanges. A common example is school choice, where a matching mechanism uses student and school preferences to assign students to schools. In such settings, forming preferences is both difficult and critical. Prior work has suggested various prediction mechanisms that help agents make decisions about their preferences. Although often deployed together, these matching and prediction mechanisms are almost always analyzed separately. The present work shows that at the intersection of the two lies a previously unexplored type of strategic behavior: agents returning to the market (e.g., schools) can attack future predictions by interacting short-term non-optimally with their matches. Here, we first introduce this type of strategic behavior, which we call an adversarial interaction attack. Next, we construct a formal economic model that captures the feedback loop between prediction mechanisms designed to assist agents and the matching mechanism used to pair them. Finally, in a simplified setting, we prove that returning agents can benefit from using adversarial interaction attacks and gain progressively more as the trust in and accuracy of predictions increases. We also show that this attack increases inequality in the student population. 

   more » « less
5. Short‐term plant–soil feedback experiment fails to predict outcome of competition observed in long‐term field experiment

   https://doi.org/10.1002/ecy.3883  

   Beckman, Noelle G. ; Dybzinski, Ray ; Tilman, David ( December 2022 , Ecology)

   Mounting evidence suggests that plant–soil feedbacks (PSF) may determine plant community structure. However, we still have a poor understanding of how predictions from short‐term PSF experiments compare with outcomes of long‐term field experiments involving competing plants. We conducted a reciprocal greenhouse experiment to examine how the growth of prairie grass species depended on the soil communities cultured by conspecific or heterospecific plant species in the field. The source soil came from monocultures in a long‐term competition experiment (LTCE; Cedar Creek Ecosystem Science Reserve, MN, USA). Within the LTCE, six species of perennial prairie grasses were grown in monocultures or in eight pairwise competition plots for 12 years under conditions of low or high soil nitrogen availability. In six cases, one species clearly excluded the other; in two cases, the pair appeared to coexist. In year 15, we gathered soil from all 12 soil types (monocultures of six species by two nitrogen levels) and grew seedlings of all six species in each soil type for 7 weeks. Using biomass estimates from this greenhouse experiment, we predicted coexistence or competitive exclusion using pairwise PSFs, as derived by Bever and colleagues, and compared model predictions to observed outcomes within the LTCE. Pairwise PSFs among the species pairs ranged from negative, which is predicted to promote coexistence, to positive, which is predicted to promote competitive exclusion. However, these short‐term PSF predictions bore no systematic resemblance to the actual outcomes of competition observed in the LTCE. Other forces may have more strongly influenced the competitive interactions or critical assumptions that underlie the PSF predictions may not have been met. Importantly, the pairwise PSF score derived by Bever et al. is only valid when the two species exhibit an internal equilibrium, corresponding to the Lotka–Volterra competition outcomes of stable coexistence and founder control. Predicting the other two scenarios, competitive exclusion by either species irrespective of initial conditions, requires measuring biomass in uncultured soil, which is methodologically challenging. Subject to several caveats that we discuss, our results call into question whether long‐term competitive outcomes in the field can be predicted from the results of short‐term PSF experiments.

    

   more » « less