City Research Online

Flatland: a Lightweight First-Person 2-D Environment for Reinforcement Learning

Caselles-Dupré, H., Annabi, L., Hagen, O., Garcia Ortiz, M. ORCID: 0000-0003-4729-7457 and Filliat, D. (2018). Flatland: a Lightweight First-Person 2-D Environment for Reinforcement Learning. Paper presented at the Workshop on Continual Unsupervised Sensorimotor Learning - ICDL-Epirob 2018, 17 September 2018, Tokyo, Japan.

Abstract

Flatlandis a simple, lightweight environment for fastprototyping and testing of reinforcement learning agents. It is oflower complexity compared to similar 3D platforms (e.g. Deep-Mind Lab or VizDoom), but emulates physical properties of thereal world, such as continuity, multi-modal partially-observablestates with first-person view and coherent physics. We proposeto use it as an intermediary benchmark for problems related toLifelong Learning.Flatlandis highly customizable and offers awide range of task difficulty to extensively evaluate the propertiesof artificial agents. We experiment with three reinforcementlearning baseline agents and show that they can rapidly solvea navigation task inFlatland. A video of an agent acting inFlatlandis available here: https://youtu.be/I5y6Y2ZypdA.

Publication Type: Conference or Workshop Item (Paper)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments: School of Mathematics, Computer Science & Engineering > Computer Science
URI: http://openaccess.city.ac.uk/id/eprint/22450
[img]
Preview
Text - Accepted Version
Download (419kB) | Preview

Export

Downloads

Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login