Learning Lattice Quantum Field Theories with Equivariant Continuous Flows

Autor: Gerdes, Mathis, de Haan, Pim, Rainone, Corrado, Bondesan, Roberto, Cheng, Miranda C. N.
Rok vydání: 2022
Předmět:
Zdroj: SciPost Phys. 15, 238 (2023)
Druh dokumentu: Working Paper
DOI: 10.21468/SciPostPhys.15.6.238
Popis: We propose a novel machine learning method for sampling from the high-dimensional probability distributions of Lattice Field Theories, which is based on a single neural ODE layer and incorporates the full symmetries of the problem. We test our model on the $\phi^4$ theory, showing that it systematically outperforms previously proposed flow-based methods in sampling efficiency, and the improvement is especially pronounced for larger lattices. Furthermore, we demonstrate that our model can learn a continuous family of theories at once, and the results of learning can be transferred to larger lattices. Such generalizations further accentuate the advantages of machine learning methods.
Comment: 17 pages, 9 figures, 1 table; slightly expanded published version, added 2 figures and 2 sections to appendix
Databáze: arXiv