Strategic Prediction with Latent Aggregative Games
This work addresses the challenge of predicting outcomes in settings with significant strategic interactions, such as voting, by providing a novel game-theoretic framework, though it appears incremental in building on existing structured prediction and game theory concepts.
The paper tackles the problem of modeling strategic interactions in structured prediction by introducing a new class of context-dependent, incomplete information games, and demonstrates that these models can recover meaningful strategic interactions from real voting data.
We introduce a new class of context dependent, incomplete information games to serve as structured prediction models for settings with significant strategic interactions. Our games map the input context to outcomes by first condensing the input into private player types that specify the utilities, weighted interactions, as well as the initial strategies for the players. The game is played over multiple rounds where players respond to weighted aggregates of their neighbors' strategies. The predicted output from the model is a mixed strategy profile (a near-Nash equilibrium) and each observation is thought of as a sample from this strategy profile. We introduce two new aggregator paradigms with provably convergent game dynamics, and characterize the conditions under which our games are identifiable from data. Our games can be parameterized in a transferable manner so that the sets of players can change from one game to another. We demonstrate empirically that our games as models can recover meaningful strategic interactions from real voting data.