# Reproducibility of fixed-node diffusion Monte Carlo across diverse community codes: The case of water–methane dimer https://mdr.nims.go.jp/datasets/96a554da-c70c-4322-ab03-17a1f0555b16 ## File - [2501.12950v3.pdf](https://mdr.nims.go.jp/filesets/6ed4dcad-5b84-4b5f-b478-a41c7327e4b4/download) ([Detail](https://mdr.nims.go.jp/filesets/6ed4dcad-5b84-4b5f-b478-a41c7327e4b4.md)) ## Id 96a554da-c70c-4322-ab03-17a1f0555b16 ## Local identifier ## Visibility open_to_public ## State published ## Created at 2025-09-10T00:54:26.824467Z ## Updated at 2025-12-01T23:30:07.623540Z ## Published at 2025-12-01T23:23:29.102261Z ## Doi ## First published url https://doi.org/10.1063/5.0272974 ## Date published 2025-09-14 ## Recorded date published 2025-9-14 ## Resource type journal_article ## Manuscript type accepted_manuscript ## Collection ## Title - title: 'Reproducibility of fixed-node diffusion Monte Carlo across diverse community codes: The case of water–methane dimer' title_type: original lang: en ## Description - description: "Fixed-node diffusion quantum Monte Carlo (FN-DMC) is a widely-trusted manybody\r\nmethod for solving the Schrödinger equation, known for its reliable predictions\r\nof material and molecular properties. Furthermore, its excellent scalability\r\nwith system complexity and near-perfect utilization of computational power makes\r\nFN-DMC ideally positioned to leverage new advances in computing to address increasingly\r\ncomplex scientific problems. Even though the method is widely used as\r\na computational gold standard, reproducibility across the numerous FN-DMC code\r\nimplementations has yet to be demonstrated. This difficulty stems from the diverse\r\narray of DMC algorithms and trial wave functions, compounded by the method’s\r\ninherent stochastic nature. This study represents a community-wide effort to assess\r\nthe reproducibility of the method, affirming that: Yes, FN-DMC is reproducible\r\n(when handled with care). Using the water-methane dimer as the canonical test\r\ncase, we compare results from eleven different FN-DMC codes and show that the\r\napproximations to treat the non-locality of pseudopotentials are the primary source\r\nof the discrepancies between them. In particular, we demonstrate that, for the same\r\nchoice of determinantal component in the trial wave function, reliable and reproducible\r\npredictions can be achieved by employing the T-move (TM), the determinant\r\nlocality approximation (DLA), or the determinant T-move (DTM) schemes, while\r\nthe older locality approximation (LA) leads to considerable variability in results.\r\nThese findings demonstrate that, with appropriate choices of algorithmic details,\r\nfixed-node DMC is reproducible across diverse community codes—highlighting the\r\nmaturity and robustness of the method as a tool for open and reliable computational\r\nscience." description_type: abstract lang: und ## Creator - name: Flaviano Della Pia role: author - name: Benjamin X. Shi role: author - name: Yasmine S. Al-Hamdani role: author - name: Dario Alfé role: author - name: Tyler A. Anderson role: author - name: Matteo Barborini role: author - name: Anouar Benali role: author - name: Michele Casula role: author - name: Neil D. Drummond role: author - name: Matúš Dubecký role: author - name: Claudia Filippi role: author - name: Paul R. C. Kent role: author - name: Jaron T. Krogel role: author - name: Pablo López Ríos role: author - name: Arne Lüchow role: author - name: Ye Luo role: author - name: Angelos Michaelides role: author - name: Lubos Mitas role: author - name: Kousuke Nakano role: author orcid: https://orcid.org/0000-0001-7756-4355 organization: National Institute for Materials Science - name: Richard J. Needs role: author - name: Manolo C. Per role: author - name: Anthony Scemama role: author - name: Jil Schultze role: author - name: Ravindra Shinde role: author - name: Emiel Slootman role: author - name: Sandro Sorella role: author - name: Alexandre Tkatchenko role: author - name: Mike Towler role: author - name: C. J. Umrigar role: author - name: Lucas K. Wagner role: author - name: William A. Wheeler role: author - name: Haihan Zhou role: author - name: Andrea Zen role: author ## Contact agent ## Publisher organization: AIP Publishing ## Managing organization ## Keyword - subject: Quantum Monte Carlo schema: not_defined - subject: Diffusion Monte Carlo schema: not_defined ## Rights - description: "Copyright 2025 Author(s). This article is distributed under a Creative Commons Attribution (CC BY) License.\r\n" identifier: https://creativecommons.org/licenses/by/4.0/ ## Other identifier(s) ## Data origin - data_origin_type: other ## Embargo ## Journal - title: The Journal of Chemical Physics issn: '00219606' volume: '163' issue: '10' article_number: '104110' ## Conference ## Related item ## Funding - identifier: CZ.10.03.01/00/22_003/0000003 funder_name: European Union Under the LERCO Project - funder_name: Operational Program Just Transition - funder_name: U.S. Department of Energy - funder_name: Basic Energy Sciences - funder_name: Computational Materials Science Program - funder_name: Center for Predictive Simulation of Functional Materials - identifier: DMR-2316007 funder_name: U.S. National Science Foundation - identifier: '1931258' funder_name: U.S. National Science Foundation - funder_name: European Center of Excellence in Exascale Computing TREX - identifier: '952165' funder_name: HORIZON EUROPE European Research Council - funder_name: JSPS Overseas Research Fellowships - identifier: JPMXS0320220025 funder_name: MEXT Leading Initiative for Excellent Young Researchers - identifier: 20222FXZ33 funder_name: European Union Under the Next Generation EU - identifier: P2022MC742 funder_name: European Union Under the Next Generation EU - identifier: RPG-2020-038 funder_name: Leverhulme Trust - identifier: '101071937' funder_name: European Research Council - identifier: FA9550-18-1-0095 funder_name: Air Force Office of Scientific Research - identifier: DE-AC05-00OR22725 funder_name: Office of Science of the U.S. Department of Energy - identifier: '90140' funder_name: IT4Innovations National Supercomputing Center - funder_name: HPC Facilities of the University of Luxembourg - identifier: A0150906493 funder_name: French Computational Resources at the CEA-TGCC Center Through the GENCI Allocation - funder_name: National Energy Research Scientific Computing Center - funder_name: Dutch National Supercomputer Snellius - funder_name: Numerical Materials Simulator at National Institute for Materials Science - identifier: 17-SC-20-SC funder_name: Exascale Computing Project - identifier: EP/T022159/1 funder_name: Engineering and Physical Sciences Research Council - identifier: EP/P020259/1 funder_name: Engineering and Physical Sciences Research Council - funder_name: DiRAC Funding From the Science and Technology Facilities Council - identifier: EP/ F036884/1 funder_name: United Kingdom Car Parrinelloconsortium ## Instrument ## Instrument operator ## Instrument managing organization ## Measurement method ## Specimen ## Chemical composition ## Structure for specimen ## Structural feature for specimen ## Specific property for specimen ## Process for specimen treatment ## Computational method ## Energy level/transition state ## Software ## Custom property ## Fileset - id: 6ed4dcad-5b84-4b5f-b478-a41c7327e4b4 filename: 2501.12950v3.pdf content_type: application/pdf size: 9820572 md5: d547abc36e360c0ab6d967a3217cc233 ## Thumbnail fileset_id: 6ed4dcad-5b84-4b5f-b478-a41c7327e4b4 filename: 2501.12950v3.pdf