An official website of the United States government
Abstract The next generation of searches for neutrinoless double beta decay ($$0 \nu \beta \beta $$ ) are poised to answer deep questions on the nature of neutrinos and the source of the Universe’s matter–antimatter asymmetry. They will be looking for event rates of less than one event per ton of instrumented isotope per year. To claim discovery, accurate and efficient simulations of detector events that mimic$$0 \nu \beta \beta $$ is critical. Traditional Monte Carlo (MC) simulations can be supplemented by machine-learning-based generative models. This work describes the performance of generative models that we designed for monolithic liquid scintillator detectors like KamLAND to produce accurate simulation data without a predefined physics model. We present their current ability to recover low-level features and perform interpolation. In the future, the results of these generative models can be used to improve event classification and background rejection by providing high-quality abundant generated data.
more »
« less
This content will become publicly available on December 1, 2026
Physics-informed machine learning for automatic model reduction in chemical reaction networks
Abstract Physics-informed machine learning bridges the gap between the high fidelity of mechanistic models and the adaptive insights of artificial intelligence. In chemical reaction network modeling, this synergy proves valuable, addressing the high computational costs of detailed mechanistic models while leveraging the predictive power of machine learning. This study applies this fusion to the biomedical challenge of A$$\beta$$fibril aggregation, a key factor in Alzheimer’s disease. Central to the research is the introduction of an automatic reaction order model reduction framework, designed to optimize reduced-order kinetic models. This framework represents a shift in model construction, automatically determining the appropriate level of detail for reaction network modeling. The proposed approach significantly improves simulation efficiency and accuracy, particularly in systems like A$$\beta$$aggregation, where precise modeling of nucleation and growth kinetics can reveal potential therapeutic targets. Additionally, the automatic model reduction technique has the potential to generalize to other network models. The methodology offers a scalable and adaptable tool for applications beyond biomedical research. Its ability to dynamically adjust model complexity based on system-specific needs ensures that models remain both computationally feasible and scientifically relevant, accommodating new data and evolving understandings of complex phenomena.
more »
« less
- Award ID(s):
- 2316003
- PAR ID:
- 10645369
- Publisher / Repository:
- Springer Nature
- Date Published:
- Journal Name:
- Scientific Reports
- Volume:
- 15
- Issue:
- 1
- ISSN:
- 2045-2322
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Emergent symmetries in atomic nuclei: Probing nuclear dynamics and physics beyond the standard modelDominant shapes naturally emerge in atomic nuclei from first principles, thereby establishing the shape-preserving symplectic Sp(3,\mathbb{R} ) symmetry as remarkably ubiquitous and almost perfect symmetry in nuclei. We discuss the critical role of this emergent symmetry in enabling machine-learning descriptions of heavy nuclei, ab initio modeling of\alpha clustering and collectivity, as well as tests of beyond-the-standard-model physics. In addition, the Sp(3,\mathbb{R} ) and SU(3) symmetries provide relevant degrees of freedom that underpin the ab initio symmetry-adapted no-core shell model with the remarkable capability of reaching nuclei and reaction fragments beyond the lightest and close-to-spherical species.more » « less
-
Abstract The question of whether the hyper-dissipative (HD) Navier-Stokes (NS) system can exhibit spontaneous formation of singularities in the super-critical regime–the hyperviscous effects being represented by a fractional power of the Laplacian, say$$\beta $$ , confined to interval$$\bigl (1, \frac{5}{4}\bigr )$$ –has been a major open problem in the mathematical fluid dynamics since the foundational work of J.L. Lions in 1960s. In this work, an evidence of criticality of the Laplacian is presented, more precisely, a class of plausible blow-up scenarios is ruled out as soon as$$\beta $$ is greater than one. While the framework is based on the ‘scale of sparseness’ of the super-level sets of the positive and negative parts of the components of the higher-order derivatives of the velocity previously introduced by the authors, a major novelty in the current work is classification of the HD flows near a potential spatiotemporal singularity in two main categories, ‘homogeneous’ (the case consistent with a near-steady behavior) and ‘non-homogenous’ (the case consistent with the formation and decay of turbulence). The main theorem states that in the non-homogeneous case any$$\beta $$ greater than one prevents a singularity. In order to illustrate the impact of this result in a methodology-free setting, a two-parameter family of dynamically rescaled blow-up profiles is considered, and it is shown that as soon as$$\beta $$ is greater than one, a new region in the parameter space is ruled out. More importantly, the region is a neighborhood (in the parameter space) of the self-similar profile, i.e., the approximately self-similar blow-up, a prime suspect in possible singularity formation, is ruled out for all HD NS models.more » « less
-
Abstract We report the identification of metastable isomeric states of$$^{228}$$ Ac at 6.28 keV, 6.67 keV and 20.19 keV, with lifetimes of an order of 100 ns. These states are produced by the$$\beta $$ -decay of$$^{228}$$ Ra, a component of the$$^{232}$$ Th decay chain, with$$\beta $$ Q-values of 39.52 keV, 39.13 keV and 25.61 keV, respectively. Due to the low Q-value of$$^{228}$$ Ra as well as the relative abundance of$$^{232}$$ Th and their progeny in low background experiments, these observations potentially impact the low-energy background modeling of dark matter search experiments.more » « less
-
Abstract Theoretical models of galaxy formation and evolution are primarily investigated through cosmological simulations and semi-analytical models. The former method consumes core-hours explicitly modeling the dynamics of the galaxies, whereas the latter method only requires core-hours foregoing directly simulating internal structure for computational efficiency. In this work, we present a proof-of-concept machine learning regression model, using a graph neural network architecture, to predict the stellar mass of high-redshift galaxies solely from their dark matter merger trees, trained from a radiation hydrodynamics cosmological simulation of the first galaxies.more » « less
