Machine Learning for Dynamic Incentive Problems

43 Pages Posted: 13 Nov 2018 Last revised: 24 Feb 2020

See all articles by Philipp Renner

Philipp Renner

Lancaster University

Simon Scheidegger

University of Lausanne - School of Economics and Business Administration (HEC-Lausanne)

Date Written: November 11, 2018


We propose a generic computational framework for solving large-scale infinite-horizon, discrete-time dynamic incentive problems with persistent hidden types. First, we combine set-valued dynamic programming techniques with unsupervised machine learning to determine irregularly-shaped feasible sets. Second, we generate training data from these pre-computed feasible sets to recursively solve the dynamic incentive problem by applying supervised and reinforcement machine learning. Third, to speed up the time-to-solution by orders of magnitude, we propose a generic parallelization scheme for dynamic incentive problems that allows for an efficient use of contemporary high-performance computing hardware. This combination enables us to analyze models of complexity that were previously considered to be intractable. To demonstrate the broad applicability of our method, we solve two very different types of dynamic incentive models: first, an adverse selection model with many discrete types; second, a problem with infinitely many types and multiple state variables.

Keywords: Dynamic Contracts, Principal-Agent Model, Dynamic Programming, Machine Learning, Gaussian Processes, High-Performance Computing

JEL Classification: C61, C73, D82, D86, E61

Suggested Citation

Renner, Philipp Johannes and Scheidegger, Simon, Machine Learning for Dynamic Incentive Problems (November 11, 2018). Available at SSRN: or

Philipp Johannes Renner

Lancaster University ( email )

Managment School
Department of Economics
Lancaster LA1 4YX, Lancashire LA1 4YX
United Kingdom

Simon Scheidegger (Contact Author)

University of Lausanne - School of Economics and Business Administration (HEC-Lausanne) ( email )

Unil Dorigny, Batiment Internef
Lausanne, 1015

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics