Database of scalable training of neural network potentials for complex interfaces through data augmentation

JSON Export

{
  "created": "2025-03-30T18:03:32.109740+00:00", 
  "id": "2619", 
  "updated": "2025-04-02T16:03:38.014449+00:00", 
  "metadata": {
    "mcid": "2025.51", 
    "_oai": {
      "id": "oai:materialscloud.org:2619"
    }, 
    "doi": "10.24435/materialscloud:w6-9a", 
    "conceptrecid": "2618", 
    "contributors": [
      {
        "familyname": "Yeu", 
        "givennames": "In Won", 
        "affiliations": [
          "Department of Chemical Engineering, Columbia University, New York, NY, USA", 
          "Columbia Center for Computational Electrochemistry, Columbia University, New York, NY, USA"
        ], 
        "email": "iy2185@columbia.edu"
      }, 
      {
        "familyname": "Stuke", 
        "givennames": "Annika", 
        "affiliations": [
          "Department of Chemical Engineering, Columbia University, New York, NY, USA", 
          "Columbia Center for Computational Electrochemistry, Columbia University, New York, NY, USA"
        ], 
        "email": "as6394@columbia.edu"
      }, 
      {
        "familyname": "Urban", 
        "givennames": "Alexander", 
        "affiliations": [
          "Department of Chemical Engineering, Columbia University, New York, NY, USA", 
          "Columbia Center for Computational Electrochemistry, Columbia University, New York, NY, USA", 
          "Columbia Electrochemical Energy Center, Columbia University, New York, NY, USA"
        ], 
        "email": "au2229@columbia.edu"
      }, 
      {
        "familyname": "Artrith", 
        "givennames": "Nongnuch", 
        "affiliations": [
          "Columbia Center for Computational Electrochemistry, Columbia University, New York, NY, USA", 
          "Debye Institute for Nanomaterials Science, Utrecht University, 3584 CS Utrecht, The Netherlands"
        ], 
        "email": "n.artrith@uu.nl"
      }
    ], 
    "license": "Creative Commons Attribution 4.0 International", 
    "is_last": true, 
    "edited_by": 576, 
    "title": "Database of scalable training of neural network potentials for complex interfaces through data augmentation", 
    "publication_date": "Apr 02, 2025, 18:03:37", 
    "references": [
      {
        "comment": "Preprint where the data is discussed", 
        "doi": "10.48550/arXiv.2412.05773", 
        "url": "https://arxiv.org/abs/2412.05773", 
        "citation": "IW Yeu, A Stuke, J L\u00f3pez-Zorrilla, JM Stevenson, DR Reichman, RA Friesner, A Urban, N Artrith, arXiv:2412.05773", 
        "type": "Preprint"
      }
    ], 
    "id": "2619", 
    "license_addendum": null, 
    "status": "published", 
    "owner": 1721, 
    "_files": [
      {
        "description": "Description of the dataset", 
        "size": 15393500, 
        "key": "README.txt", 
        "checksum": "md5:02fd7bde8f16958144845121ba163d86"
      }, 
      {
        "description": "ANN training data in XSF format used for H\u2082 molecule example (Total number: 403)", 
        "size": 12697, 
        "key": "1_H2_xsf.tar.bz2", 
        "checksum": "md5:0aa065a05906aabb4a8d5c1fd992e50a"
      }, 
      {
        "description": "ANN training data in XSF format used for EC dimer example (Total number: 181,000)", 
        "size": 160212576, 
        "key": "2_EC-EC_xsf.tar.bz2", 
        "checksum": "md5:12c48fdeabada0f3483e296397c811ba"
      }, 
      {
        "description": "ANN training data in XSF format used for EC on Li surface example (Total number: 68,000)", 
        "size": 134492502, 
        "key": "3_Li-EC-surface_xsf.tar.bz2", 
        "checksum": "md5:6a1ce0f80ef1bf7899eb2a831656d8b4"
      }, 
      {
        "description": "ANN training data in XSF format used for Li-EC interface example (Total number: 80,768)", 
        "size": 155118490, 
        "key": "4_Li-EC-interfaces_xsf.tar.bz2", 
        "checksum": "md5:05f3542df4fa980c62dad03f01456a23"
      }
    ], 
    "description": "This database contains the reference data used for direct force training of Artificial Neural Network (ANN) interatomic potentials using the atomic energy network (\u00e6net) and \u00e6net-PyTorch packages (https://github.com/atomisticnet/aenet-PyTorch). It also includes the GPR-augmented data used for indirect force training via Gaussian Process Regression (GPR) surrogate models using the \u00e6net-GPR package (https://github.com/atomisticnet/aenet-gpr).\nEach data file contains atomic structures, energies, and atomic forces in XCrySDen Structure Format (XSF). The dataset includes all reference training/test data and corresponding GPR-augmented data used in the four benchmark examples presented in the reference paper, \u201cScalable Training of Neural Network Potentials for Complex Interfaces Through Data Augmentation\u201d.\nA hierarchy of the dataset is described in the README.txt file, and an overview of the dataset is also summarized in supplementary Table S1 of the reference paper.", 
    "version": 1, 
    "keywords": [
      "first principles", 
      "machine learning", 
      "Li metal battery", 
      "potential energy surface", 
      "aenet", 
      "data augmentation"
    ]
  }, 
  "revision": 10
}