2026/05/07 by Tatsuhiro Misumi
Physics and Astronomy · #Quantum many-body systems #Physics of Superconductivity and Magnetism #Quantum Chromodynamics and Particle Interactions
paper · pdf · doi:10.1103/756m-stz5
We propose a novel, machine-learning-based framework for constructing lattice fermions using physics-informed neural networks (PINNs). Our approach treats the formulation of the Dirac operator as an optimization problem guided by physical requirements, such as symmetries, locality and doubler-decoupling conditions. We first demonstrate that, when trained to satisfy the Ginsparg-Wilson (GW) relation as a soft constraint, a neural network reproduces the overlap fermion operator to high numerical accuracy and learns an effective sign-function mapping without explicitly using a prescribed polynomial or rational approximation. Second, we extend the framework from operator construction to machine-assisted algebraic discovery. Within a generalized polynomial ansatz, the network autonomously drives higher-order terms to zero and recovers the standard Ginsparg-Wilson relation. Remarkably, by changing the initial search bias, the same framework also finds a distinct solution corresponding to a Fujikawa-type generalized GW relation.