OCAtari: Object-Centric Atari 2600 Reinforcement Learning Environments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Delfosse, Quentin, Blüml, Jannis, Gregori, Bjarne, Sztwiertnia, Sebastian, Kersting, Kristian
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914692658626560
author Delfosse, Quentin
Blüml, Jannis
Gregori, Bjarne
Sztwiertnia, Sebastian
Kersting, Kristian
author_facet Delfosse, Quentin
Blüml, Jannis
Gregori, Bjarne
Sztwiertnia, Sebastian
Kersting, Kristian
contents Cognitive science and psychology suggest that object-centric representations of complex scenes are a promising step towards enabling efficient abstract reasoning from low-level perceptual features. Yet, most deep reinforcement learning approaches only rely on pixel-based representations that do not capture the compositional properties of natural scenes. For this, we need environments and datasets that allow us to work and evaluate object-centric approaches. In our work, we extend the Atari Learning Environments, the most-used evaluation framework for deep RL approaches, by introducing OCAtari, that performs resource-efficient extractions of the object-centric states for these games. Our framework allows for object discovery, object representation learning, as well as object-centric RL. We evaluate OCAtari's detection capabilities and resource efficiency. Our source code is available at github.com/k4ntz/OC_Atari.
format Preprint
id arxiv_https___arxiv_org_abs_2306_08649
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle OCAtari: Object-Centric Atari 2600 Reinforcement Learning Environments
Delfosse, Quentin
Blüml, Jannis
Gregori, Bjarne
Sztwiertnia, Sebastian
Kersting, Kristian
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Cognitive science and psychology suggest that object-centric representations of complex scenes are a promising step towards enabling efficient abstract reasoning from low-level perceptual features. Yet, most deep reinforcement learning approaches only rely on pixel-based representations that do not capture the compositional properties of natural scenes. For this, we need environments and datasets that allow us to work and evaluate object-centric approaches. In our work, we extend the Atari Learning Environments, the most-used evaluation framework for deep RL approaches, by introducing OCAtari, that performs resource-efficient extractions of the object-centric states for these games. Our framework allows for object discovery, object representation learning, as well as object-centric RL. We evaluate OCAtari's detection capabilities and resource efficiency. Our source code is available at github.com/k4ntz/OC_Atari.
title OCAtari: Object-Centric Atari 2600 Reinforcement Learning Environments
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2306.08649