arXiv · 2305.16774
How to Understand Limitations of Generative Networks
Abstract
Well-trained classifiers and their complete weight distributions provide us with a well-motivated and practicable method to test generative networks in particle physics. We illustrate their benefits for distribution-shifted jets, calorimeter showers, and reconstruction-level events. In all cases, the classifier weights make for a powerful test of the generative network, identify potential problems in the density estimation, relate them to the underlying physics, and tie in with a comprehensive precision and uncertainty treatment for generative networks.
Explore related subjects
Keep this discovery
Ranit Das, Luigi Favaro, Theo Heimel, Claudius Krause, Tilman Plehn, David Shih. 2023-05-26. How to Understand Limitations of Generative Networks. https://doi.org/10.21468/scipostphys.16.1.031
Cite the original work for its findings. Save a collection to share your selection of sources.