Related papers: Optimization of Nuclear Mass Models Using Algorith…
Interpreting neural networks is a crucial and challenging task in machine learning. In this paper, we develop a novel framework for detecting statistical interactions captured by a feedforward multilayer neural network by directly…
Network quantization has rapidly become one of the most widely used methods to compress and accelerate deep neural networks. Recent efforts propose to quantize weights and activations from different layers with different precision to…
Neutrino physics has entered an era of precision measurements. With these precise measurements, we may be able to distinguish different models that have been constructed to explain the small neutrino masses and the large mixing among them.…
The mechanism of resonance enhancement of neutrino oscillations in matter and some critical remarks to this mechanism are considered. Using of this resonance mechanism is very important to examine the model of electroweak interactions since…
For light nuclei, ab initio many-body methods such as the no-core shell model are the tools of choice for predictive, high-precision nuclear structure calculations. The applicability and the level of precision of these methods, however, is…
Machine learning potentials have become an important tool for atomistic simulations in many fields, from chemistry via molecular biology to materials science. Most of the established methods, however, rely on local properties and are thus…
Neural network quantum states (NQS) have emerged as a powerful and flexible framework for addressing quantum many-body problems. While successful for model Hamiltonians, their application to molecular systems remains challenging for several…
The accuracy of three different sets of Hartree-Fock-Bogoliubov calculations of nuclear binding energies is systematically evaluated. To emphasize minor fluctuations, a second order, four-point mass relation, which almost completely…
Our picture of the electroweak interactions continues to improve, with ever more precise constraints on the masses of the Higgs boson(s) and on non-standard physics. Some recent developments include: (a) a calculation of higher-order…
This paper proposes a novel binarized weight network (BT) for a resource-efficient neural structure. The proposed model estimates a binary representation of weights by taking into account the approximation error with an additional term.…
The possible application of boosted neural network to particle classification in high energy physics is discussed. A two-dimensional toy model, where the boundary between signal and background is irregular but not overlapping, is…
We investigate the correlation between the neutrino mass limit and dark energy with time evolving equation of state. Parameterizing dark energy as w=w_0+w_1*z/(1+z), we make a global fit using Markov Chain Monte Carlo technique to determine…
Optimizing deep neural networks (DNNs) often suffers from the ill-conditioned problem. We observe that the scaling-based weight space symmetry property in rectified nonlinear network will cause this negative effect. Therefore, we propose to…
Data-driven, machine learning (ML) models of atomistic interactions are often based on flexible and non-physical functions that can relate nuanced aspects of atomic arrangements into predictions of energies and forces. As a result, these…
Three versions of the quark-meson coupling (QMC) model are applied to describe properties of nuclear matter and finite nuclei. The models differ in the treatment of the bag constant and in terms of nonlinear scalar self-interactions. As a…
We have constructed an empirical formulae for the fusion and interaction barriers using experimental values available till date. The fusion barriers so obtained have been compared with different model predictions based on the proximity,…
Semi-empirical mass formula of the atomic nucleus describe binding energies of the nuclei. In the simple form of this formula, there are five terms related to the properties of the nuclear structure. The coefficients in each terms can be…
Calculations of nuclear masses, using nuclear density functional theory, are presented for even-even nuclei spanning the nuclear chart. The resulting binding energy differences can be interpreted in terms of valence proton-neutron…
Properties of nuclear and neutron matter are discussed in a nonlinear $\sigma$-$\omega$-$\rho$ mean-field approximation with self-interactions and mixing-interactions of mesons and baryons. The nonlinear interactions are renormalized by…
We consider the problem of training a neural network to store a set of patterns with maximal noise robustness. A solution, in terms of optimal weights and state update rules, is derived by training each individual neuron to perform either…