Abstract
The particle-in-cell numerical method of plasma physics balances a trade-off between computational cost and intrinsic noise. Inference on data produced by these simulations generally consists of binning the data to recover the particle distribution function, from which physical processes may be investigated. In addition to containing noise, the distribution function is temporally dynamic and can be non-gaussian and multi-modal, making the task of modeling it difficult. Here we demonstrate the use of normalizing flows to learn a smooth, tractable approximation to the noisy particle distribution function. We demonstrate that the resulting data driven likelihood conserves relevant physics and may be extended to encapsulate the temporal evolution of the distribution function.
Partially Supported by Department of Energy Grant 17-SC-20-SC.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Agostinelli, F., Anderson, M.R., Lee, H.: Adaptive multi-column deep neural networks with application to robust image denoising. In: Proceedings of the 26th International Conference on Neural Information Processing Systems, NIPS 2013, vol. 1, pp. 1493–1501. Curran Associates Inc., Red Hook (2013)
Bigdeli, S.A., Lin, G., Portenier, T., Dunbar, L.A., Zwicker, M.: Learning generative models using denoising density estimators (2020)
Block, A., Mroueh, Y., Rakhlin, A.: Generative modeling with denoising auto-encoders and Langevin sampling (2020)
Cho, K.: Simple sparsification improves sparse denoising autoencoders in denoising highly noisy images. In: 30th International Conference on Machine Learning, ICML 2013, 16 June 2013–21 June 2013, pp. 1469–1477 (2013)
Choi, J.Y., et al.: Coupling exascale multiphysics applications: methods and lessons learned, pp. 442–452 (2018). https://doi.org/10.1109/eScience.2018.00133
Dilokthanakul, N., et al.: Deep unsupervised clustering with Gaussian mixture variational autoencoders (2017)
Dinh, L., Krueger, D., Bengio, Y.: NICE: non-linear independent components estimation (2015)
Dinh, L., Sohl-Dickstein, J., Bengio, S.: Density estimation using real NVP (2017)
Dominski, J., et al.: Spatial coupling of gyrokinetic simulations, a generalized scheme based on first-principles. Phys. Plasmas 28(2) (2021). https://doi.org/10.1063/5.0027160
Dominski, J., et al.: A tight-coupling scheme sharing minimum information across a spatial interface between gyrokinetic turbulence codes. Phys. Plasmas 25(7), 072308 (2018). https://doi.org/10.1063/1.5044707
Durkan, C., Bekasov, A., Murray, I., Papamakarios, G.: nflows: normalizing flows in PyTorch (November 2020). https://doi.org/10.5281/zenodo.4296287
Germaschewski, K., et al.: The plasma simulation code: a modern particle-in-cell code with patch-based load-balancing. J. Comput. Phys. 318, 305–326 (2016). https://doi.org/10.1016/j.jcp.2016.05.013. https://www.sciencedirect.com/science/article/pii/S0021999116301413
Goodfellow, I.J., et al.: Generative adversarial networks (2014)
Kingma, D.P., Dhariwal, P.: Glow: generative flow with invertible 1 \(\times \) 1 convolutions (2018)
Kingma, D.P., Welling, M.: Auto-encoding variational Bayes (2014)
Lezhnin, K.V., et al.: Kinetic simulations of electron pre-energization by magnetized collisionless shocks in expanding laboratory plasmas. Astrophys. J. 908(2), L52 (2021). https://doi.org/10.3847/2041-8213/abe407
Merlo, G., et al.: First coupled GENE–XGC microturbulence simulations. Phys. Plasmas 28(1), 012303 (2021). https://doi.org/10.1063/5.0026661
Papamakarios, G., Pavlakou, T., Murray, I.: Masked autoregressive flow for density estimation (2018)
Rezende, D.J., Mohamed, S.: Variational inference with normalizing flows (2016)
Rivero, A.D., Dvorkin, C.: Flow-based likelihoods for non-gaussian inference. Phys. Rev. D 102(10) (2020). https://doi.org/10.1103/physrevd.102.103507
Xie, J., Xu, L., Chen, E.: Image denoising and inpainting with deep neural networks. In: Advances in Neural Information Processing Systems, vol. 25. Curran Associates, Inc. (2012). https://proceedings.neurips.cc/paper/2012/file/6cdd60ea0045eb7a6ec44c54d29ed402-Paper.pdf
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Donaghy, J., Germaschewski, K. (2022). Inference and De-noising of Non-gaussian Particle Distribution Functions: A Generative Modeling Approach. In: Nicosia, G., et al. Machine Learning, Optimization, and Data Science. LOD 2021. Lecture Notes in Computer Science(), vol 13163. Springer, Cham. https://doi.org/10.1007/978-3-030-95467-3_25
Download citation
DOI: https://doi.org/10.1007/978-3-030-95467-3_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-95466-6
Online ISBN: 978-3-030-95467-3
eBook Packages: Computer ScienceComputer Science (R0)