تقرير
AdvNF: Reducing Mode Collapse in Conditional Normalising Flows using Adversarial Learning
| العنوان: | AdvNF: Reducing Mode Collapse in Conditional Normalising Flows using Adversarial Learning |
|---|---|
| المؤلفون: | Kanaujia, Vikas, Scheurer, Mathias S., Arora, Vipul |
| المصدر: | SciPost Phys. 16, 132 (2024) |
| سنة النشر: | 2024 |
| المجموعة: | Computer Science Condensed Matter Physics (Other) |
| مصطلحات موضوعية: | Computer Science - Machine Learning, Condensed Matter - Statistical Mechanics, Physics - Computational Physics |
| الوصف: | Deep generative models complement Markov-chain-Monte-Carlo methods for efficiently sampling from high-dimensional distributions. Among these methods, explicit generators, such as Normalising Flows (NFs), in combination with the Metropolis Hastings algorithm have been extensively applied to get unbiased samples from target distributions. We systematically study central problems in conditional NFs, such as high variance, mode collapse and data efficiency. We propose adversarial training for NFs to ameliorate these problems. Experiments are conducted with low-dimensional synthetic datasets and XY spin models in two spatial dimensions. Comment: 29 pages, submitted to Scipost Physics |
| نوع الوثيقة: | Working Paper |
| DOI: | 10.21468/SciPostPhys.16.5.132 |
| URL الوصول: | http://arxiv.org/abs/2401.15948 |
| رقم الأكسشن: | edsarx.2401.15948 |
| قاعدة البيانات: | arXiv |
| DOI: | 10.21468/SciPostPhys.16.5.132 |
|---|