Learning Game-Theoretic Models of Multiagent Trajectories Using Implicit Layers

TL;DR

This paper introduces an end-to-end neural architecture that combines game-theoretic reasoning with implicit layers to predict multiagent trajectories, providing interpretability and transferability to decision-making tasks.

Contribution

It presents a novel hybrid neural network with differentiable game-theoretic components, including a new class of continuous potential games and equilibrium refinement, enabling interpretable trajectory prediction.

Findings

Accurately predicts highway merging trajectories in real-world data

Demonstrates transferability to decision-making tasks

Provides theoretical guarantees for gradient computation and model soundness

Abstract

For prediction of interacting agents' trajectories, we propose an end-to-end trainable architecture that hybridizes neural nets with game-theoretic reasoning, has interpretable intermediate representations, and transfers to downstream decision making. It uses a net that reveals preferences from the agents' past joint trajectory, and a differentiable implicit layer that maps these preferences to local Nash equilibria, forming the modes of the predicted future trajectory. Additionally, it learns an equilibrium refinement concept. For tractability, we introduce a new class of continuous potential games and an equilibrium-separating partition of the action space. We provide theoretical results for explicit gradients and soundness. In experiments, we evaluate our approach on two real-world data sets, where we predict highway driver merging trajectories, and on a simple decision-making transfer task.