Converting sWeights to probabilities with density ratios
- Thomas Jefferson National Accelerator Facility (TJNAF), Newport News, VA (United States)
The use of machine learning approaches continues to have many benefits in experimental nuclear and particle physics. One common issue is generating training data which is sufficiently realistic to give reliable results. Here we advocate using real experimental data as the source of training data and demonstrate how one might subtract background contributions through the use of probabilistic weights which can be readily applied to training data. The sPlot formalism is a common tool used to isolate distributions from different sources. However, the negative sWeights produced by the sPlot technique can cause training problems and poor predictive power. This article demonstrates how density ratio estimation can be applied to convert sWeights to event probabilities, which we call drWeights. The drWeights can then be applied to produce the distributions of interest and are consistent with direct use of the sWeights. This article will also show how decision trees are particularly well suited to convert sWeights, with the benefit of fast prediction rates and adaptability to aspects of experimental data such as the data sample size and proportions of different event sources. We also show that a density ratio product approach in which the initial drWeights are reweighted by an additional converter gives substantially better results.
- Research Organization:
- Thomas Jefferson National Accelerator Facility (TJNAF)
- Sponsoring Organization:
- US Department of Energy Office of Nuclear Energy; Science and Technology Facilities Council; U.S. Department of Energy; Office of Science; Nuclear Physics
- Grant/Contract Number:
- AC05-06OR23177
- OSTI ID:
- 2999561
- Report Number(s):
- JLAB-PHY-24-4183; arXiv:2409.08183; DOE/OR/23177-7669
- Journal Information:
- Computer Physics Communications, Journal Name: Computer Physics Communications Vol. 318; ISSN 0010-4655
- Publisher:
- Elsevier BVCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Similar Records
Pairwise Association of Seismic Arrivals with Convolutional Neural Networks
Semisupervised Learning for Seismic Monitoring Applications