Physics-Informed Discriminator (PID) for Conditional Generative Adversarial Nets
| dc.contributor.author | Daw, Arka | en |
| dc.contributor.author | Maruf, M. | en |
| dc.contributor.author | Karpatne, Anuj | en |
| dc.contributor.department | Computer Science | en |
| dc.date.accessioned | 2021-04-26T16:40:49Z | en |
| dc.date.available | 2021-04-26T16:40:49Z | en |
| dc.date.issued | 2020-12-11 | en |
| dc.description.abstract | We propose a novel method of incorporating physical knowledge as an additional input to the discriminator of a conditional Generative Adversarial Net (cGAN). Our proposed approach, termed as Physics-informed Discriminator for cGAN (cGAN-PID), is more aligned to the adversarial learning idea of cGAN as opposed to existing methods on incorporating physical knowledge in GANs by adding physics based loss functions as additional terms in the optimization objective of GAN. We evaluate the performance of our model on two toy datasets and demonstrate that our proposed cGAN-PID can be used as an alternative to the existing techniques. | en |
| dc.format.mimetype | application/pdf | en |
| dc.identifier.uri | http://hdl.handle.net/10919/103125 | en |
| dc.language.iso | en | en |
| dc.publisher | NeurIPS | en |
| dc.relation.ispartof | Machine Learning and Physical Sciences workshop at Neural Information Processing Systems (NeurIPS) 2020. | en |
| dc.rights | Attribution 4.0 International | en |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | en |
| dc.subject | Machine learning | en |
| dc.subject | Generative Adversarial Networks | en |
| dc.subject | Physics Informed Neural Networks | en |
| dc.subject | uncertainty quantification | en |
| dc.subject | theory-guided data science | en |
| dc.title | Physics-Informed Discriminator (PID) for Conditional Generative Adversarial Nets | en |
| dc.type | Conference proceeding | en |
| dc.type.dcmitype | Text | en |