{"@context":"https://schema.org","@type":"CreativeWork","@id":"https://forgecascade.org/public/capsules/b5fd2114-39c9-4127-94c0-2dd70104c521","identifier":"b5fd2114-39c9-4127-94c0-2dd70104c521","url":"https://forgecascade.org/public/capsules/b5fd2114-39c9-4127-94c0-2dd70104c521","name":"Geometric regularization of autoencoders via observed stochastic dynamics","text":"# Geometric regularization of autoencoders via observed stochastic dynamics\n\nSource-backed public reference for autoencoders, stochastic dynamics, manifold learning.\n\nSummary: The paper studies reduced simulators for stochastic dynamical systems and adds geometry-aware penalties to autoencoder training. It uses observed ambient covariance to constrain tangent-bundle behavior and improve downstream latent dynamics.\n\nKey points:\n- Targets slow or metastable stochastic systems represented in high-dimensional space.\n- Uses tangent-bundle and inverse-consistency penalties in a chart-learning pipeline.\n- Analyzes how chart error propagates into learned drift, diffusion, and ambient dynamics.\n\nPublic review note: Source-backed ML/math reference suitable as a free technical capsule.\n\nSource: https://arxiv.org/abs/2604.16282\nAuthors: Sean Hill, Felix X.-F. Ye\nPublished: 2026-04-17","keywords":["machine-learning","autoencoders","stochastic-dynamics","manifold-learning"],"about":[],"citation":[],"isPartOf":{"@type":"Dataset","name":"Forge Cascade Knowledge Graph","url":"https://forgecascade.org"},"publisher":{"@type":"Organization","name":"Forge Cascade","url":"https://forgecascade.org"},"dateCreated":"2026-05-15T03:35:32.353060Z","dateModified":"2026-06-19T01:59:49.343691Z","isBasedOn":"https://arxiv.org/abs/2604.16282","additionalProperty":[{"@type":"PropertyValue","name":"trust_level","value":95},{"@type":"PropertyValue","name":"verification_status","value":"sources_verified"},{"@type":"PropertyValue","name":"provenance_status","value":"valid"},{"@type":"PropertyValue","name":"evidence_level","value":"primary_source"},{"@type":"PropertyValue","name":"content_hash","value":"65bef390074ac8a9113a1d3c6fa8ade15d9eb4bdc848d05e571dee5c82dfb7c5"}]}