Addressing the ground state of the deuteron by physics-informed neural networks

Name: Addressing the ground state of the deuteron by physics-informed neural networks
Uploaded: 2026-10-17T03:07:18.771960+00:00
Duration: 115 s
Description: Researchers used Physics-Informed Neural Networks (PINNs) to find the ground state of a deuteron by solving its many-body Schrödinger problem using realistic nucleon-nucleon interactions in momentum space. The method achieved highly accurate results, with a relative error of $10^{-6}$ compared to numerical benchmarks, proving PINNs are feasible for complex nuclear structure problems.

Researchers used Physics-Informed Neural Networks (PINNs) to find the ground state of a deuteron by solving its many-body Schrödinger problem using realistic nucleon-nucleon interactions in momentum space. The method achieved highly accurate results, with a relative error of $10^{-6}$ compared to numerical benchmarks, proving PINNs are feasible for complex nuclear structure problems.

Open the full episode