The exploration of complex physical phenomena requires precise and rigorous comparison of theoretical model calculations and experimental measurements in order to validate the physical picture underlying the models, discover new effects, and to determine the parameters that best describe the phenomena. This characterizes an “Inverse Problem,” in which causal factors are deduced from measurements that are influenced by them. Inverse problems are crucial to a wide variety of scientific and engineering disciplines. In particular, they play a central role in experimentation and theory/data comparisons for many areas of modern Nuclear Physics (NP), High-Energy Physics (HEP), and Cosmology.
Bayes’s Theorem is a powerful tool to solve Inverse Problems, providing conceptually transparent and unbiased constraints on theoretical parameters and their uncertainties (“Bayesian Inference”), and enabling the quantification of agreement or tension between models and data. However, analyses based on Bayesian Inference are often challenging for NP and HEP applications, either because of the large number of parameters in the problem, the high computational cost of the calculations, or both.
Bayesian inference methods can require significant computing resources; for example, inference involving many model parameters in the phase space, or inference with a computationally-expensive forward model. Machine learning (ML)-based approaches are essential to avoid brute-force and computationally intensive calculations for such studies. The various projects in this proposal are connected by this common thread.
The BUQ project is a multi-institutional collaboration for the development and deployment of novel Bayesian analysis tools to advance the scientific scope of a broad range of current and future NP experiments. This project brings together NP domain scientists working on several high-profile NP projects for which new, high-performance Bayesian Uncertainty Quantification (“Bayesian UQ”) methods are essential to carry out the science, with data scientists who are developing forefront methods that are applicable to these problems. The NP projects in this proposal comprise measurements of the mass and fundamental nature of the neutrino; study of the Quark-Gluon Plasma that filled the early universe; and mapping of natural and anthropogenic radiation environments. These projects all address high-priority topics in the DOE OS/NP portfolio, as presented in the 2015 NSAC Long Range Plan. The methods developed in this project will however be more widely applicable, thereby also advancing science in the larger NP portfolio.