Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.19913 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914216735145984 |
|---|---|
| author | Drnevich, Matthew Jiggins, Stephen Cranmer, Kyle |
| author_facet | Drnevich, Matthew Jiggins, Stephen Cranmer, Kyle |
| contents | We consider a generalization of the classifier-based density-ratio estimation task to a quasiprobabilistic setting where probability densities can be negative. The problem with most loss functions used for this task is that they implicitly define a relationship between the optimal classifier and the target quasiprobabilistic density ratio which is discontinuous or not surjective. We address these problems by introducing a convex loss function that is well-suited for both probabilistic and quasiprobabilistic density ratio estimation. To quantify performance, an extended version of the Sliced-Wasserstein distance is introduced which is compatible with quasiprobability distributions. We demonstrate our approach on a real-world example from particle physics, of di-Higgs production in association with jets via gluon-gluon fusion, and achieve state-of-the-art results. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_19913 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Quasiprobabilistic Density Ratio Estimation with a Reverse Engineered Classification Loss Function Drnevich, Matthew Jiggins, Stephen Cranmer, Kyle Machine Learning High Energy Physics - Experiment We consider a generalization of the classifier-based density-ratio estimation task to a quasiprobabilistic setting where probability densities can be negative. The problem with most loss functions used for this task is that they implicitly define a relationship between the optimal classifier and the target quasiprobabilistic density ratio which is discontinuous or not surjective. We address these problems by introducing a convex loss function that is well-suited for both probabilistic and quasiprobabilistic density ratio estimation. To quantify performance, an extended version of the Sliced-Wasserstein distance is introduced which is compatible with quasiprobability distributions. We demonstrate our approach on a real-world example from particle physics, of di-Higgs production in association with jets via gluon-gluon fusion, and achieve state-of-the-art results. |
| title | Quasiprobabilistic Density Ratio Estimation with a Reverse Engineered Classification Loss Function |
| topic | Machine Learning High Energy Physics - Experiment |
| url | https://arxiv.org/abs/2512.19913 |