Application of neural network to sign problem via path optimization method

Abstract

We introduce the feedforward neural network to attack the sign problem via the path optimization method. The variables of integration is complexified and the integration path is optimized in the complexified space by minimizing the cost function which reflects the seriousness of the sign problem. For the preparation and optimization of the integral path in multi-dimensional systems, we utilize the feedforward neural network. We examine the validity and usefulness of the method in the two-dimensional complex $\lambda {\varphi }^4$ theory at finite chemical potential as an example of the quantum field theory having the sign problem. We show that the average phase factor is significantly enhanced after the optimization and then we can safely perform the hybrid Monte-Carlo method.

10 pages, 4 figures, v3: some comments and references added

References in corpus (2)

• Can stochastic quantization evade the sign problem? -- the relativistic Bose gas at finite chemical potential
• Deep Learning Beyond Lefschetz Thimbles

Cited by in corpus (35)

• Review on novel methods for lattice gauge theories
• Mapping neutron star data to the equation of state using the deep neural network
• Regressive and generative neural networks for scalar field theory
• Complex Langevin and other approaches to the sign problem in quantum many-body physics
• Exploring QCD matter in extreme conditions with Machine Learning
• Identifying the nature of the QCD transition in relativistic collision of heavy nuclei with deep learning
• QCD Equation of State at Finite Chemical Potentials for Relativistic Nuclear Collisions
• Tensor renormalization group approach to four-dimensional complex ${\varphi }^4$ theory at finite density
• Finite Density $QE{D}_{1+1}$ Near Lefschetz Thimbles
• Control the model sign problem via path optimization method: Monte-Carlo approach to QCD effective model with Polyakov loop
• Path integral contour deformations for observables in $SU(N)$ gauge theory
• Path integral contour deformations for noisy observables
• Normalizing Flows and the Real-Time Sign Problem
• Leveraging Machine Learning to Alleviate Hubbard Model Sign Problems
• A simple approach towards the sign problem using path optimisation
• Application of the path optimization method to the sign problem in an effective model of QCD with a repulsive vector-type interaction
• Path optimization in 0+1 dimensional QCD at finite density
• Reweighting Lefschetz Thimbles
• Mitigating the Hubbard Sign Problem with Complex-Valued Neural Networks
• Continuous-mixture Autoregressive Networks for efficient variational calculation of many-body systems
• Exponential reduction of the sign problem at finite density in the 2+1D XY model via contour deformations
• Persistent homology analysis for dense QCD effective model with heavy quarks
• Simulating Yang-Mills theories with a complex coupling
• Gauge invariant input to neural network for path optimization method
• Decoding Quantum Field Theory with Machine Learning
• Improving efficiency of the path optimization method for a gauge theory
• Stochastic quantization and diffusion models
• Color superconductivity on the lattice -- analytic predictions from QCD in a small box
• Anatomy of the dense QCD matter from canonical sectors
• Towards sampling complex actions
• Anatomy of a strong residual sign problem on the thimbles
• Single Particle Spectrum of Doped ${\mathrm{C}}_{20}{\mathrm{H}}_{12}$-Perylene
• Application of the path optimization method to a discrete spin system
• Nonanalyticity, sign problem and Polyakov line in Z3-symmetric heavy quark model at low temperature: Phenomenological model analyses
• Path optimization method for the sign problem caused by fermion determinant