Collective Stochastic Discrete Choice Problems: A Min-LQG Game Formulation

TL;DR

This paper models large-scale collective discrete choice problems as a Min-LQG game, deriving decentralized strategies via mean field game theory that predict the probabilistic distribution of choices in large populations.

Contribution

It introduces a novel Min-LQG game framework for collective choice, providing explicit solutions and convergence results for decentralized strategies in large populations.

Findings

Decentralized strategies converge to Nash equilibrium as population grows.

The model predicts the probabilistic distribution of choices among agents.

Explicit solutions for individual best responses are derived.

Abstract

We consider a class of dynamic collective choice models with social interactions, whereby a large number of non-uniform agents have to individually settle on one of multiple discrete alternative choices, with the relevance of their would-be choices continuously impacted by noise and the unfolding group behavior. This class of problems is modeled here as a so-called Min-LQG game, i.e., a linear quadratic Gaussian dynamic and non-cooperative game, with an additional combinatorial aspect in that it includes a final choice-related minimization in its terminal cost. The presence of this minimization term is key to enforcing some specific discrete choice by each individual agent. The theory of mean field games is invoked to generate a class of decentralized agent feedback control strategies which are then shown to converge to an exact Nash equilibrium of the game as the number of players increases to infinity. A key building block in our approach is an explicit solution to the problem of computing the best response of a generic agent to some arbitrarily posited smooth mean field trajectory. Ultimately, an agent is shown to face a continuously revised discrete choice problem, where greedy choices dictated by current conditions must be constantly balanced against the risk of the future process noise upsetting the wisdom of such decisions.Even though an agent's ultimately chosen alternative is random and dictated by its entire noise history and initial state, the limiting infinite population macroscopic behavior can still be predicted. It is shown that any Nash equilibrium of the game is defined by an a priori computable probability matrix characterizing the manner in which the agent population ultimately splits among the available alternatives.