Physics-informed Neural Networks with Unknown Measurement Noise
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2022
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909227299110912 |
|---|---|
| author | Pilar, Philipp Wahlström, Niklas |
| author_facet | Pilar, Philipp Wahlström, Niklas |
| contents | Physics-informed neural networks (PINNs) constitute a flexible approach to both finding solutions and identifying parameters of partial differential equations. Most works on the topic assume noiseless data, or data contaminated with weak Gaussian noise. We show that the standard PINN framework breaks down in case of non-Gaussian noise. We give a way of resolving this fundamental issue and we propose to jointly train an energy-based model (EBM) to learn the correct noise distribution. We illustrate the improved performance of our approach using multiple examples. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2211_15498 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Physics-informed Neural Networks with Unknown Measurement Noise Pilar, Philipp Wahlström, Niklas Machine Learning Physics-informed neural networks (PINNs) constitute a flexible approach to both finding solutions and identifying parameters of partial differential equations. Most works on the topic assume noiseless data, or data contaminated with weak Gaussian noise. We show that the standard PINN framework breaks down in case of non-Gaussian noise. We give a way of resolving this fundamental issue and we propose to jointly train an energy-based model (EBM) to learn the correct noise distribution. We illustrate the improved performance of our approach using multiple examples. |
| title | Physics-informed Neural Networks with Unknown Measurement Noise |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2211.15498 |