Simple, Good, Fast: Self-Supervised World Models Free of Baggage

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Robine, Jan, Höftmann, Marc, Harmeling, Stefan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912411235123200
author Robine, Jan
Höftmann, Marc
Harmeling, Stefan
author_facet Robine, Jan
Höftmann, Marc
Harmeling, Stefan
contents What are the essential components of world models? How far do we get with world models that are not employing RNNs, transformers, discrete representations, and image reconstructions? This paper introduces SGF, a Simple, Good, and Fast world model that uses self-supervised representation learning, captures short-time dependencies through frame and action stacking, and enhances robustness against model errors through data augmentation. We extensively discuss SGF's connections to established world models, evaluate the building blocks in ablation studies, and demonstrate good performance through quantitative comparisons on the Atari 100k benchmark.
format Preprint
id arxiv_https___arxiv_org_abs_2506_02612
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Simple, Good, Fast: Self-Supervised World Models Free of Baggage
Robine, Jan
Höftmann, Marc
Harmeling, Stefan
Machine Learning
Artificial Intelligence
What are the essential components of world models? How far do we get with world models that are not employing RNNs, transformers, discrete representations, and image reconstructions? This paper introduces SGF, a Simple, Good, and Fast world model that uses self-supervised representation learning, captures short-time dependencies through frame and action stacking, and enhances robustness against model errors through data augmentation. We extensively discuss SGF's connections to established world models, evaluate the building blocks in ablation studies, and demonstrate good performance through quantitative comparisons on the Atari 100k benchmark.
title Simple, Good, Fast: Self-Supervised World Models Free of Baggage
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.02612