Abstract:Joint-Embedding Predictive Architectures (JEPAs) provide a simpleframework for learning world models by predicting future latent this http URL, JEPA training is subject to a bias-variance this http URL sufficient structural constraints, excessive representationalvariance causes the model to collapse to trivial this http URL recent LeWorldModel (LeWM) shows that this issue can be alleviated bysimply constraining latent embeddings with an isotropic Gaussian this http URL, latent representations inherently lie on low-dimensional manifoldswithin a high-dimensional ambient space, and enforcing an isotropic Gaussianprior directly in this ambient space introduces an overly strong this http URL this work, we propose ame, which seeks a favorable operatingpoint on the bias-variance frontier by applying Gaussian constraints inmultiple random subspaces rather than in the originalembedding this http URL design relaxes the global constraint while preserving itsanti-collapse effect, leading to a better balance between trainingstability and representation this http URL experiments across fourcontinuous-control environments demonstrate that consistentlyoutperforms LeWM with very clear this http URL method is simple yet effective, and serves as a strong baseline for future JEPA-based world model this http URL code is available at this https URL.
| Comments: | this https URL |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.09241 [cs.LG] |
| (or arXiv:2605.09241v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.09241 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Kai Zhao [view email]
[v1]
Sun, 10 May 2026 00:51:47 UTC (756 KB)