Providing physics guides in Bayesian neural networks from input layer: case of giant dipole resonance predictions

Abstract

The Bayesian neural network (BNN) has been applied to evaluate and predict the nuclear data. However, how to provide physics guides in BNN is a key but an open question. In this work, the case study on giant dipole resonance (GDR) energy is presented to illustrate the effectiveness and maneuverability of the method to provide physics guides in BNN from input layer. The Spearman's correlation coefficients are applied to assess the statistical dependence between nuclear properties in the ground state and the GDR energies. Then the optimal ground-state properties are employed as the input layer in the BNN for evaluating and predicting the GDR energies. Those selected ground-state properties actively contributes to reduce the predicted errors and avoid the risk of the non-physics divergence. This work gives a demonstration to find effects of the GDR energy by using the BNN without the physics motivated model, which may be helpful for discovering physics effects from the complex nuclear data.

References in corpus (19)

• Weight Uncertainty in Neural Networks
• Surface diffuseness correction in global mass formula
• Markov Chain Monte Carlo and Variational Inference: Bridging the Gap
• Nuclear mass predictions based on Bayesian neural network approach with pairing and shell effects
• Nuclear Mass Predictions for the Crustal Composition of Neutron Stars: A Bayesian Neural Network Approach
• Bayesian approach to model-based extrapolation of nuclear observables
• Nuclear charge radii: Density functional theory meets Bayesian neural networks
• Giant Dipole Resonance as a Fingerprint of $\alpha$ Clustering Configurations in ${}^{12}$C and ${}^{16}$O
• Symmetry energy from the nuclear collective motion: constraints from dipole, quadrupole, monopole and spin-dipole resonances
• Bayesian Evaluation of Incomplete Fission Yields
• Nuclear mass systematics using neural networks
• Giant Dipole Resonance Parameters of Ground-State Photoabsorption: Experimental Values with Uncertainties
• Systematics on ground-state energies of nuclei within the neural networks
• Decoding Beta-Decay Systematics: A Global Statistical Model for Beta^- Halflives
• The unrestricted Skyrme-tensor time-dependent Hartree-Fock and its application to the nuclear response from spherical to triaxial nuclei
• Constraining simultaneously nuclear symmetry energy and neutron-proton effective mass splitting with nucleus giant resonances from a dynamical approach
• The description of giant dipole resonance key parameters with multitask neural networks
• Constraining the in-medium nucleon-nucleon cross section from the width of nuclear giant dipole resonance
• Giant dipole resonance and shape evolution in Nd isotopes within TDHF method

Cited by in corpus (6)

• Multi-task learning on nuclear masses and separation energies with the kernel ridge regression
• Prediction of Nuclear Charge Density Distribution with Feedback Neural Network
• Determining impact parameters of heavy-ion collisions at low-intermediate incident energies using deep learning with convolutional neural network
• Inference of Parameters for Back-shifted Fermi Gas Model using Feedback Neural Network
• Constraining neutron-proton effective mass splitting through nuclear giant dipole resonance within transport approach
• Further exploration of the machine-learning-based nuclear mass table