Enregistré dans:
Détails bibliographiques
Auteurs principaux: Hauri, Michael, Zenke, Friedemann
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2603.07083
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911590803046400
author Hauri, Michael
Zenke, Friedemann
author_facet Hauri, Michael
Zenke, Friedemann
contents Model-based reinforcement learning (MBRL) agents operating in high-dimensional observation spaces, such as Dreamer, rely on learning abstract representations for effective planning and control. Existing approaches typically employ reconstruction-based objectives in the observation space, which can render representations sensitive to task-irrelevant details. Recent alternatives trade reconstruction for auxiliary action prediction heads or view augmentation strategies, but perform worse in the Crafter environment than reconstruction-based methods. We close this gap between Dreamer and reconstruction-free models by introducing a JEPA-style predictor defined on continuous, deterministic representations. Our method matches Dreamer's performance on Crafter, demonstrating effective world model learning on this benchmark without reconstruction objectives.
format Preprint
id arxiv_https___arxiv_org_abs_2603_07083
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Dreamer-CDP: Improving Reconstruction-free World Models Via Continuous Deterministic Representation Prediction
Hauri, Michael
Zenke, Friedemann
Machine Learning
Model-based reinforcement learning (MBRL) agents operating in high-dimensional observation spaces, such as Dreamer, rely on learning abstract representations for effective planning and control. Existing approaches typically employ reconstruction-based objectives in the observation space, which can render representations sensitive to task-irrelevant details. Recent alternatives trade reconstruction for auxiliary action prediction heads or view augmentation strategies, but perform worse in the Crafter environment than reconstruction-based methods. We close this gap between Dreamer and reconstruction-free models by introducing a JEPA-style predictor defined on continuous, deterministic representations. Our method matches Dreamer's performance on Crafter, demonstrating effective world model learning on this benchmark without reconstruction objectives.
title Dreamer-CDP: Improving Reconstruction-free World Models Via Continuous Deterministic Representation Prediction
topic Machine Learning
url https://arxiv.org/abs/2603.07083