InvertibleNetworks.jl: A Julia package for scalable normalizing flows
| Title | InvertibleNetworks.jl: A Julia package for scalable normalizing flows |
| Publication Type | Journal Article |
| Year of Publication | 2024 |
| Authors | Rafael Orozco, Philipp A. Witte, Mathias Louboutin, Ali Siahkoohi, Gabrio Rizzuti, Bas Peters, Felix J. Herrmann |
| Journal | Journal of Open Source Software |
| Volume | 9 |
| Month | 7 |
| Keywords | Bayesian inference, computing, conditional normalizing flows, deep learning, HPC, Inverse problems, memory, Normalizing flows, software, Uncertainty quantification |
| Abstract | InvertibleNetworks.jl is a Julia package designed for the scalable implementation of normalizing flows, a method for density estimation and sampling in high-dimensional distributions. This package excels in memory efficiency by leveraging the inherent invertibility of normalizing flows, which significantly reduces memory requirements during backpropagation compared to existing normalizing flow packages that rely on automatic differentiation frameworks. InvertibleNetworks.jl has been adapted for diverse applications, including seismic imaging, medical imaging, and CO2 monitoring, demonstrating its effectiveness in learning high-dimensional distributions. |
| URL | https://doi.org/10.21105/joss.06554 |
| DOI | 10.21105/joss.06554 |
| Citation Key | orozco2023invnet |