A Survey of Reinforcement Learning For Economics

TL;DR

This survey explores how reinforcement learning extends dynamic programming to complex economic models with high-dimensional states and strategic interactions, highlighting its potential and current limitations.

Contribution

It reviews the connection between classical planning and modern reinforcement learning algorithms in economics, demonstrating their mechanics and practical challenges.

Findings

Reinforcement learning extends tractability to high-dimensional economic problems.

Algorithms show brittleness and sample inefficiency outside tabular settings.

RL's success depends on accurate simulators and economic structure guidance.

Abstract

This survey (re)introduces reinforcement learning methods to economists. The curse of dimensionality limits how far exact dynamic programming can be effectively applied, forcing us to rely on suitably "small" problems or our ability to convert "big" problems into smaller ones. While this reduction has been sufficient for many classical applications, a growing class of economic models resists such reduction. Reinforcement learning algorithms offer a natural, sample-based extension of dynamic programming, extending tractability to problems with high-dimensional states, continuous actions, and strategic interactions. I review the theory connecting classical planning to modern learning algorithms and demonstrate their mechanics through simulated examples in pricing, inventory control, strategic games, and preference elicitation. I also examine the practical vulnerabilities of these algorithms, noting their brittleness, sample inefficiency, sensitivity to hyperparameters, and the absence of global convergence guarantees outside of tabular settings. The successes of reinforcement learning remain strictly bounded by these constraints, as well as a reliance on accurate simulators. When guided by economic structure, reinforcement learning provides a remarkably flexible framework. It stands as an imperfect, but promising, addition to the computational economist's toolkit. A companion survey (Rust and Rawat, 2026b) covers the inverse problem of inferring preferences from observed behavior. All simulation code is publicly available.