journal article·2024-11

Multi‐Task Mixture Density Graph Neural Networks for Predicting Catalyst Performance

Chen Liang, Bowen Wang, Shaogang Hao, Guangyong Chen, Pheng‐Ann Heng, Xiaolong Zou

Advanced Functional Materials· Volume 34· Issue 45
Publisher page

Abstract

Abstract Graph neural networks (GNNs) have drawn more and more attention from material scientists and demonstrated a strong capacity to establish connections between structures and properties. However, with only unrelaxed structures provided as input, few GNN models can predict the thermodynamic properties of relaxed configurations with an acceptable level of error. In this work, a multi‐task (MT) architecture based on DimeNet++ and mixture density networks is developed to improve the performance of such task. Taking CO adsorption on Cu‐based single‐atom alloy catalysts as an example, the method can reliably predict CO adsorption energy with a mean absolute error of 0.087 eV from the initial CO adsorption structures without costly first‐principles calculations. Compared to other state‐of‐the‐art GNN methods, the model exhibits improved generalization ability when predicting the catalytic performance of out‐of‐distribution configurations, built with either unseen substrate surfaces or doping species. Further, the enhancement of expressivity has also been demonstrated on the IS2RE predicting task in the Open Catalyst 2020 project. The proposed MT GNN strategy can facilitate the catalyst discovery and optimization process.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Machine Learning in Materials Science · Catalytic Processes in Materials Science · Catalysis and Oxidation Reactions

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Shenzhen Geim Graphene Center & Shenzhen Key Laboratory of Advanced Layered Materials for Value‐added Applications Institute of Materials Research Tsinghua Shenzhen International Graduate School Tsinghua University Shenzhen 518055 China

Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong 999077 China

Tencent Shenzhen 518054 China

Zhejiang Lab Zhejiang University Hangzhou 311121 China

Funding

National Natural Science Foundation of China · Guangdong Innovative and Entrepreneurial Research Team Program

References (94)

Show source-provided references
1. 10.1021/cr300463y10.1021/cr300463y ↗
2. 10.1016/j.apsusc.2023.15856010.1016/j.apsusc.2023.158560 ↗
3. 10.1038/ncomms1543810.1038/ncomms15438 ↗
4. 10.1039/c0ee00071j10.1039/c0ee00071j ↗
5. 10.1002/anie.20120832010.1002/anie.201208320 ↗
6. 10.1021/jacs.0c0498110.1021/jacs.0c04981 ↗
7. 10.1038/s41929-018-0142-110.1038/s41929-018-0142-1 ↗
8. 10.1021/jz201461p10.1021/jz201461p ↗
9. 10.1038/s41586-018-0337-210.1038/s41586-018-0337-2 ↗
10. 10.1038/s41524-021-00514-810.1038/s41524-021-00514-8 ↗
11. 10.1021/acs.jpclett.2c0047210.1021/acs.jpclett.2c00472 ↗
12. 10.1038/s41586-020-2242-810.1038/s41586-020-2242-8 ↗
13. 10.1039/C9TA13404B10.1039/C9TA13404B ↗
14. 10.1038/s41929-018-0063-z10.1038/s41929-018-0063-z ↗
15. 10.1021/jacs.8b0880010.1021/jacs.8b08800 ↗
16. 10.1039/D0TA06203K10.1039/D0TA06203K ↗
17. 10.1021/acs.jpcc.0c0596410.1021/acs.jpcc.0c05964 ↗
18. 10.1016/j.gee.2021.10.00310.1016/j.gee.2021.10.003 ↗
19. 10.1039/D1TA09184K10.1039/D1TA09184K ↗
20. 10.1021/acs.jpclett.5b0166010.1021/acs.jpclett.5b01660 ↗
21. 10.1021/jacs.9b0495610.1021/jacs.9b04956 ↗
22. 10.1109/IJCNN.2005.155594210.1109/IJCNN.2005.1555942 ↗
23. 10.1109/TNN.2008.200560510.1109/TNN.2008.2005605 ↗
24. Defferrard M. · 2016
25. T. N.Kipf M.Welling inInt. Conf. on Learn. Represent. 2017.
26. 10.1038/s41524-019-0221-010.1038/s41524-019-0221-0 ↗
27. 10.1021/acs.jpclett.1c0329110.1021/acs.jpclett.1c03291 ↗
28. 10.1002/adma.20200265810.1002/adma.202002658 ↗
29. Trinajstic N. · 1992
30. Gilmer J. · 2017 · 1263
31. 10.1021/acs.jpclett.2c0092910.1021/acs.jpclett.2c00929 ↗
32. 10.1038/s41467-020-14969-810.1038/s41467-020-14969-8 ↗
33. 10.1021/acscatal.2c0367510.1021/acscatal.2c03675 ↗
34. D. K.Duvenaud D.Maclaurin J.Iparraguirre R.Bombarell T.Hirzel A.Aspuru‐Guzik R. P.Adams inConvolutional Networks on Graphs for Learning Molecular Fingerprints NIPS Red Hook NY 2015.
35. 10.1038/ncomms1389010.1038/ncomms13890 ↗
36. 10.1063/1.501977910.1063/1.5019779 ↗
37. 10.1063/1.501118110.1063/1.5011181 ↗
38. 10.1103/PhysRevLett.120.14530110.1103/PhysRevLett.120.145301 ↗
39. 10.1021/acs.chemmater.9b0129410.1021/acs.chemmater.9b01294 ↗
40. 10.1021/acs.jctc.9b0018110.1021/acs.jctc.9b00181 ↗
41. 10.1021/acs.jmedchem.9b0095910.1021/acs.jmedchem.9b00959 ↗
42. J.Klicpera J.Groß S.Günnemann inInt. Conf. on Learn. Represent. 2020.
43. J.Klicpera S.Giri J. T.Margraf S.Günnemann inMach. Learn. for Mol. Workshop at NeurIPS 2020.
44. K.Schütt O.Unke M.Gastegger Proceedings of the 38th International Conference on Machine Learning PMLR Cambridge2021 pp.9377–9388.
45. 10.1038/s41524-022-00739-110.1038/s41524-022-00739-1 ↗
46. Klicpera J. · 2021
47. 10.1021/acs.jpclett.9b0142810.1021/acs.jpclett.9b01428 ↗
48. 10.1038/s41467-022-33256-210.1038/s41467-022-33256-2 ↗
49. 10.1038/s41524-022-00891-810.1038/s41524-022-00891-8 ↗
50. 10.1023/A:100737960673410.1023/A:1007379606734 ↗
51. 10.1038/s41467-021-25639-810.1038/s41467-021-25639-8 ↗
52. Bishop C. M. · 1994
53. 10.1038/s42256-021-00409-910.1038/s42256-021-00409-9 ↗
54. 10.1021/acscatal.6b0139310.1021/acscatal.6b01393 ↗
55. 10.1021/acscatal.8b0088110.1021/acscatal.8b00881 ↗
56. 10.1002/aenm.20190262510.1002/aenm.201902625 ↗
57. 10.1021/acs.jpcc.8b1244910.1021/acs.jpcc.8b12449 ↗
58. 10.1021/acscatal.1c0575010.1021/acscatal.1c05750 ↗
59. 10.1080/17458080.2021.195903210.1080/17458080.2021.1959032 ↗
60. 10.1038/s41467-021-25426-510.1038/s41467-021-25426-5 ↗
61. 10.1021/acs.jpclett.0c0067610.1021/acs.jpclett.0c00676 ↗
62. 10.1021/acscatal.0c0452510.1021/acscatal.0c04525 ↗
63. Anderson B. · 2019
64. V. C. G.Satorras E.Hoogeboom M.Welling inProc. of the 38th Int. Conf. on Mach. Learn. 2021 pp.9323–9332.
65. J.Brandstetter R.Hesselink E.van derPol E. J.Bekkers M.Welling inInt. Conf. on Learn. Represent. 2022.
66. 10.1038/s41467-022-29939-510.1038/s41467-022-29939-5 ↗
67. P.Thölke G. D.Fabritiis inInt. Conf. on Learn. Represent. 2022.
68. 10.1038/s41467-023-38468-810.1038/s41467-023-38468-8 ↗
69. Y.‐L.Liao T.Smidt inEleventh Int. Conf. on Learn. Represent. 2023.
70. 10.1038/s42004-023-01045-710.1038/s42004-023-01045-7 ↗
71. I.Misra A.Shrivastava A.Gupta M.Hebert inProc. of the IEEE conf. on comput. vision and pattern recognition IEEE Heidelberg2016 pp.3994–4003.10.1109/CVPR.2016.433 ↗
72. 10.1103/PhysRevLett.98.14640110.1103/PhysRevLett.98.146401 ↗
73. 10.1021/acsenergylett.8b0193310.1021/acsenergylett.8b01933 ↗
74. 10.1214/aos/101320345110.1214/aos/1013203451 ↗
75. 10.1021/acscatal.9b0189910.1021/acscatal.9b01899 ↗
76. 10.1021/acs.jpclett.5b0072210.1021/acs.jpclett.5b00722 ↗
77. 10.1021/ja302668n10.1021/ja302668n ↗
78. 10.1038/s41467-020-20615-010.1038/s41467-020-20615-0 ↗
79. 10.1073/pnas.161893511410.1073/pnas.1618935114 ↗
80. 10.1021/jacs.9b1070910.1021/jacs.9b10709 ↗
81. 10.1073/pnas.182168311710.1073/pnas.1821683117 ↗
82. 10.1039/C7SC02267K10.1039/C7SC02267K ↗
83. J.Godwin M.Schaarschmidt A. L.Gaunt A.Sanchez‐Gonzalez Y.Rubanova P.Veličković J.Kirkpatrick P.Battaglia inInt. Conf. on Learn. Represent. 2022.
84. J.Gasteiger M.Shuaibi A.Sriram S.Günnemann Z. W.Ulissi C. L.Zitnick A.Das inTransact. on Mach. Learn. Res. 2022.
85. C. L.Zitnick A.Das A.Kolluru J.Lan M.Shuaibi A.Sriram Z. W.Ulissi B. M.Wood inAdv. in Neural Informat. Process. Sys. 2022.
86. C.Ying T.Cai S.Luo S.Zheng G.Ke D.He Y.Shen T.‐Y.Liu Adv. inNeural Informat. Process. Sys.2021 34 28877.
87. Y.Shi S.Zheng G.Ke Y.Shen J.You J.He S.Luo C.Liu D.He T.‐Y.Liu arXiv preprint arXiv:2203.048102022.
88. 10.1609/aaai.v38i14.2947110.1609/aaai.v38i14.29471 ↗
89. 10.1016/0927-0256(96)00008-010.1016/0927-0256(96)00008-0 ↗
90. 10.1103/PhysRevB.59.175810.1103/PhysRevB.59.1758 ↗
91. 10.1103/PhysRevB.59.741310.1103/PhysRevB.59.7413 ↗
92. Pedregosa F. · J. Mach. Learn. Res. · 2011 · 12 · 2825
93. K.He X.Zhang S.Ren J.Sun inProc. IEEE Conf. on Computer Vision and Pattern Recognition IEEE Heidelberg2016 pp.770–778.10.1109/CVPR.2016.90 ↗
94. Paszke A. · 2019