MHG-GNN: Combination of Molecular Hypergraph Grammar with Graph Neural Network

Akihiro Kishimoto; Hiroshi Kajino; Masataka Hirose; Junta Fuchiwaki; Indra Priyadarsini; Lisa Hamada; Hajime Shinohara; Daiju Nakano; Seiji Takeda

MHG-GNN: Combination of Molecular Hypergraph Grammar with Graph Neural Network

Machine Learning 2023-09-29 v1

Authors: Akihiro Kishimoto , Hiroshi Kajino , Masataka Hirose , Junta Fuchiwaki , Indra Priyadarsini , Lisa Hamada , Hajime Shinohara , Daiju Nakano , Seiji Takeda

View on arXiv ↗ PDF ↗

Abstract

Property prediction plays an important role in material discovery. As an initial step to eventually develop a foundation model for material science, we introduce a new autoencoder called the MHG-GNN, which combines graph neural network (GNN) with Molecular Hypergraph Grammar (MHG). Results on a variety of property prediction tasks with diverse materials show that MHG-GNN is promising.

Keywords

property prediction material discovery graph neural network molecular hypergraph grammar autoencoder foundation model

Cite

@article{arxiv.2309.16374,
  title  = {MHG-GNN: Combination of Molecular Hypergraph Grammar with Graph Neural Network},
  author = {Akihiro Kishimoto and Hiroshi Kajino and Masataka Hirose and Junta Fuchiwaki and Indra Priyadarsini and Lisa Hamada and Hajime Shinohara and Daiju Nakano and Seiji Takeda},
  journal= {arXiv preprint arXiv:2309.16374},
  year   = {2023}
}

Comments

8 pages, 1 figure

MHG-GNN: Combination of Molecular Hypergraph Grammar with Graph Neural Network

Abstract

Keywords

Cite

Comments

Related papers