Obstacle Tower Without Human Demonstrations: How Far a Deep Feed-Forward Network Goes with Reinforcement Learning

IEEE Conference on Games (CoG), vol. 80, pp. 447–454

Abstract

The Obstacle Tower Challenge is the task to master a procedurally generated chain of levels that subsequently get harder to complete. Whereas the most top performing entries of last year's competition used human demonstrations or reward shaping to learn how to cope with the challenge, we present an approach that performed competitively (placed 7th) but starts completely from scratch by means of Deep Reinforcement Learning with a relatively simple feed-forward deep network structure. We especially look at the generalization performance of the taken approach concerning different seeds and various visual themes that have become available after the competition, and investigate where the agent fails and why. Note that our approach does not possess a short-term memory like employing recurrent hidden states. With this work, we hope to contribute to a better understanding of what is possible with a relatively simple, flexible solution that can be applied to learning in environments featuring complex 3D visual input where the abstract task structure itself is still fairly simple.

Authors 4

1. Marco Pleines

   TU Dortmund University

   Affiliation as printed

   Faculty of Computer Science, Technische Universität Dortmund, Dortmund, Germany

   Technical University of Dortmund,

2. Jenia Jitsev

   Forschungszentrum Jülich · Jülich Supercomputing Centre

   Affiliation as printed

   Juelich Supercomputing Center (JSC), Helmholtz Research Center Juelich, Juelich, Germany

   Forschungszentrum Julich

3. Mike Preuß Aachen

   Leiden University

   Affiliation as printed

   LIACS, Universiteit Leiden, Leiden, NL

   Leiden University

4. Fränk Zimmer

   Rhine-Waal University of Applied Sciences

   Affiliation as printed

   Faculty of Communication and Environment, Hochschule Rhein-Waal, Kamp-Lintfort, Germany

   Hochschule Rhein-Waal,Faculty of Communication and Environment,Kamp-Lintfort,Germany