Deep Incentive Design with Differentiable Equilibrium Blocks

TL;DR

This paper introduces a novel differentiable framework called deep incentive design (DID) that uses game-agnostic equilibrium modules to efficiently solve complex multi-agent incentive problems across various domains.

Contribution

The paper presents differentiable equilibrium blocks (DEBs) as a new modular approach enabling neural networks to solve diverse incentive design tasks in a unified manner.

Findings

Successfully applied to contract design, machine scheduling, and inverse equilibrium problems.

Handles a wide range of game sizes from two to sixteen actions per player.

Achieves solution of full distribution of problem instances with a single neural network.

Abstract

Automated design of multi-agent interactions with desirable equilibrium outcomes is inherently difficult due to the computational hardness, non-uniqueness, and instability of the resulting equilibria. In this work, we propose the use of game-agnostic differentiable equilibrium blocks (DEBs) as modules in a novel, differentiable framework to address a wide variety of incentive design problems from economics and computer science. We call this framework deep incentive design (DID). To validate our approach, we examine three diverse, challenging incentive design tasks: contract design, machine scheduling, and inverse equilibrium problems. For each task, we train a single neural network using a unified pipeline and DEB. This architecture solves the full distribution of problem instances, parameterized by a context, handling all games across a wide range of scales (from two to sixteen actions per player).