Volltext-Downloads (blau) und Frontdoor-Views (grau)

Iterative Oblique Decision Trees Deliver Explainable RL Models

  • The demand for explainable and transparent models increases with the continued success of reinforcement learning. In this article, we explore the potential of generating shallow decision trees (DTs) as simple and transparent surrogate models for opaque deep reinforcement learning (DRL) agents. We investigate three algorithms for generating training data for axis-parallel and oblique DTs with the help of DRL agents (“oracles”) and evaluate these methods on classic control problems from OpenAI Gym. The results show that one of our newly developed algorithms, the iterative training, outperforms traditional sampling algorithms, resulting in well-performing DTs that often even surpass the oracle from which they were trained. Even higher dimensional problems can be solved with surprisingly shallow DTs. We discuss the advantages and disadvantages of different sampling methods and insights into the decision-making process made possible by the transparent nature of DTs. Our work contributes to the development of not only powerful but also explainable RL agents and highlights the potential of DTs as a simple and effective alternative to complex DRL models.

Export metadata

Additional Services

Search Google Scholar


Author:Raphael C. Engelhardt, Marc Oedingen, Moritz Lange, Laurenz Wiskott, Wolfgang Konen
Parent Title (English):Algorithms
Editor:Mehmet Aydin, Rafet Durgut, Abdur Rakib
Document Type:Article
Date of first Publication:2023/05/31
Date of Publication (online):2023/06/29
GND-Keyword:Bestärkendes Lernen <Künstliche Intelligenz>; Entscheidungsbaum; Explainable Artificial Intelligence
Tag:Decision Tree; Explainable AI; Reinforcement Learning; Rule Learning
Page Number:24
Institutes:Informatik und Ingenieurwissenschaften (F10) / Fakultät 10 / Institut für Informatik
Dewey Decimal Classification:600 Technik, Medizin, angewandte Wissenschaften
Open Access:Open Access
Licence (German):License LogoCreative Commons - CC BY - Namensnennung 4.0 International