Unsupervised Learning of Hybrid Latent Dynamics: A Learn-to-Identify Framework
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866929274152288256 |
|---|---|
| author | Ye, Yubo Vadhavkar, Sumeet Jiang, Xiajun Missel, Ryan Liu, Huafeng Wang, Linwei |
| author_facet | Ye, Yubo Vadhavkar, Sumeet Jiang, Xiajun Missel, Ryan Liu, Huafeng Wang, Linwei |
| contents | Modern applications increasingly require unsupervised learning of latent dynamics from high-dimensional time-series. This presents a significant challenge of identifiability: many abstract latent representations may reconstruct observations, yet do they guarantee an adequate identification of the governing dynamics? This paper investigates this challenge from two angles: the use of physics inductive bias specific to the data being modeled, and a learn-to-identify strategy that separates forecasting objectives from the data used for the identification. We combine these two strategies in a novel framework for unsupervised meta-learning of hybrid latent dynamics (Meta-HyLaD) with: 1) a latent dynamic function that hybridize known mathematical expressions of prior physics with neural functions describing its unknown errors, and 2) a meta-learning formulation to learn to separately identify both components of the hybrid dynamics. Through extensive experiments on five physics and one biomedical systems, we provide strong evidence for the benefits of Meta-HyLaD to integrate rich prior knowledge while identifying their gap to observed data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_08194 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Unsupervised Learning of Hybrid Latent Dynamics: A Learn-to-Identify Framework Ye, Yubo Vadhavkar, Sumeet Jiang, Xiajun Missel, Ryan Liu, Huafeng Wang, Linwei Machine Learning Modern applications increasingly require unsupervised learning of latent dynamics from high-dimensional time-series. This presents a significant challenge of identifiability: many abstract latent representations may reconstruct observations, yet do they guarantee an adequate identification of the governing dynamics? This paper investigates this challenge from two angles: the use of physics inductive bias specific to the data being modeled, and a learn-to-identify strategy that separates forecasting objectives from the data used for the identification. We combine these two strategies in a novel framework for unsupervised meta-learning of hybrid latent dynamics (Meta-HyLaD) with: 1) a latent dynamic function that hybridize known mathematical expressions of prior physics with neural functions describing its unknown errors, and 2) a meta-learning formulation to learn to separately identify both components of the hybrid dynamics. Through extensive experiments on five physics and one biomedical systems, we provide strong evidence for the benefits of Meta-HyLaD to integrate rich prior knowledge while identifying their gap to observed data. |
| title | Unsupervised Learning of Hybrid Latent Dynamics: A Learn-to-Identify Framework |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2403.08194 |