Label Propagation Training Schemes for Physics-Informed Neural Networks and Gaussian Processes
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916197907300352 |
|---|---|
| author | Zhong, Ming Liu, Dehao Arroyave, Raymundo Braga-Neto, Ulisses |
| author_facet | Zhong, Ming Liu, Dehao Arroyave, Raymundo Braga-Neto, Ulisses |
| contents | This paper proposes a semi-supervised methodology for training physics-informed machine learning methods. This includes self-training of physics-informed neural networks and physics-informed Gaussian processes in isolation, and the integration of the two via co-training. We demonstrate via extensive numerical experiments how these methods can ameliorate the issue of propagating information forward in time, which is a common failure mode of physics-informed machine learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_05817 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Label Propagation Training Schemes for Physics-Informed Neural Networks and Gaussian Processes Zhong, Ming Liu, Dehao Arroyave, Raymundo Braga-Neto, Ulisses Machine Learning This paper proposes a semi-supervised methodology for training physics-informed machine learning methods. This includes self-training of physics-informed neural networks and physics-informed Gaussian processes in isolation, and the integration of the two via co-training. We demonstrate via extensive numerical experiments how these methods can ameliorate the issue of propagating information forward in time, which is a common failure mode of physics-informed machine learning. |
| title | Label Propagation Training Schemes for Physics-Informed Neural Networks and Gaussian Processes |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2404.05817 |