journal article·2023-07Open access

Generative artificial intelligence and its applications in materials science: Current situation and future perspectives

Yue Liu, Zhengwei Yang, Zhenyao Yu, Zitu Liu, Dahui Liu, Hailong Lin, Mingqing Li, Shuchang Ma, Maxim Avdeev, Siqi Shi

Journal of Materiomics· Volume 9· Issue 4

Abstract

Generative Artificial Intelligence (GAI) is attracting the increasing attention of materials community for its excellent capability of generating required contents. With the introduction of Prompt paradigm and reinforcement learning from human feedback (RLHF), GAI shifts from the task-specific to general pattern gradually, enabling to tackle multiple complicated tasks involved in resolving the structure-activity relationships. Here, we review the development status of GAI comprehensively and analyze pros and cons of various generative models in the view of methodology. The applications of task-specific generative models involving materials inverse design and data augmentation are also dissected. Taking ChatGPT as an example, we explore the potential applications of general GAI in generating multiple materials content, solving differential equation as well as querying materials FAQs. Furthermore, we summarize six challenges encountered for the use of GAI in materials science and provide the corresponding solutions. This work paves the way for providing effective and explainable materials data generation and analysis approaches to accelerate the materials research and development.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Machine Learning in Materials Science · Ferroelectric and Negative Capacitance Devices · Advanced Memory and Neural Computing

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

References (100+)

Show source-provided references
1. Materials discovery and design using machine learning10.1016/j.jmat.2017.08.002 ↗
2. Machine learning assisted materials design and discovery for rechargeable batteries10.1016/j.ensm.2020.06.033 ↗
3. Generative artificial intelligence: trends and prospects10.1109/MC.2022.3192720 ↗
4. A survey on generative diffusion model
5. A survey on generative adversarial networks: variants, applications, and training10.1145/3463475 ↗
6. Inverse molecular design using machine learning: generative models for matter engineering10.1126/science.aat2663 ↗
7. Predicting molecular conformation via dynamic graph score matching
8. Generative models for inverse design of inorganic solid materials
9. Equivariant diffusion for molecule generation in 3D
10. Physics guided deep learning for generative design of crystal materials with symmetry constraints10.1038/s41524-023-00987-9 ↗
11. Rick · 2020
12. Pre-train, prompt, and predict: a systematic survey of prompting methods in natural language processing
13. Generalizing from a few examples: a survey on few-shot learning10.1145/3386252 ↗
14. Lampert, Bernt Schiele, and Zeynep Akata. Zero-shot learning—a comprehensive evaluation of the good, the bad and the ugly10.1109/TPAMI.2018.2857768 ↗
15. Training language models to follow instructions with human feedback
16. Sparks of artificial general intelligence: early experiments with GPT-4
17. Chengwei Qin, Aston Zhang, Zhuosheng Zhang, Jiaao Chen, Michihiro Yasunaga, and Diyi Yang. Is chatgpt a general-purpose natural language processing task solver?. arXiv:230206476.
18. Generative adversarial networks10.1145/3422622 ↗
19. Mehdi Mirza and Simon Osindero. Conditional generative adversarial nets. arXiv:14111784.
20. Alec Radford, Luke Metz, and Soumith Chintala. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv:151106434.
21. Wasserstein generative adversarial networks
22. Improved training of wasserstein GANs
23. Efros. Unpaired image-to-image translation using cycle-consistent adversarial networks
24. Noseong Park, Mahmoud Mohammadi, Kshitij Gorde, Sushil Jajodia, Hongkyu Park, and Youngmin Kim. Data synthesis based on generative adversarial networks. arXiv:180603384.
25. Lei Xu and Kalyan Veeramachaneni. Synthesizing tabular data using generative adversarial networks. arXiv:181111264.
26. Modeling tabular data using conditional GAN
27. Data augmentation in microscopic images for material data mining
28. Predicting ultra-high-performance concrete compressive strength using tabular generative adversarial networks10.3390/ma13214757 ↗
29. A two-step data augmentation method based on generative adversarial network for hardness prediction of high entropy alloy10.1016/j.commatsci.2023.112064 ↗
30. Generative adversarial networks (GAN) based efficient sampling of chemical composition space for inverse design of inorganic materials10.1038/s41524-020-00352-0 ↗
31. Computational discovery of new 2D materials using deep learning generative models10.1021/acsami.1c01044 ↗
32. Generative adversarial networks for crystal structure prediction10.1021/acscentsci.0c00426 ↗
33. Constrained crystals deep convolutional generative adversarial network for the inverse design of crystal structures10.1038/s41524-021-00526-4 ↗
34. Diederik P Kingma and Max Welling. Auto-encoding variational bayes. arXiv:13126114.
35. An introduction to autoencoders
36. Autoencoding beyond pixels using a learned similarity metric
37. Variational lossy autoencoder
38. NVAE: a deep hierarchical variational autoencoder
39. Neural discrete representation learning
40. Learning disentangled joint continuous and discrete representations
41. Generating diverse high-fidelity images with vq-vae-2
42. Higgins · 2016
43. Disentangling by factorising
44. Isolating sources of disentanglement in variational autoencoders
45. Automatic chemical design using a data-driven continuous representation of molecules10.1021/acscentsci.7b00572 ↗
46. Inverse design of solid-state materials via a continuous representation10.1016/j.matt.2019.08.017 ↗
47. TextureVAE: learning interpretable representations of material microstructures using variational autoencoders
48. A binded VAE for inorganic material generation
49. Denoising diffusion probabilistic models
50. Diffusion models beat GANs on image synthesis
51. Denoising diffusion implicit models
52. Learning to efficiently sample from diffusion probabilistic models
53. Learning fast samplers for diffusion models by differentiating through sample quality
54. GENIE: higher-order denoising diffusion solvers
55. On distillation of guided diffusion models
56. Score-based generative modeling in latent space
57. Structured denoising diffusion models in discrete state-spaces
58. Improved denoising diffusion probabilistic models
59. Variational diffusion models
60. Maximum likelihood training of score-based diffusion models
61. Analytic-dpm: an analytic estimate of the optimal reverse variance in diffusion probabilistic models
62. Microstructure design of multifunctional particulate composite materials using conditional diffusion models
63. Protein structure and sequence generation with equivariant denoising diffusion probabilistic models
64. Structure-based drug design with equivariant diffusion models
65. Protein sequence and structure co-design with equivariant translation
66. Nice: non-linear independent components estimation
67. Laurent Dinh, Jascha Sohl-Dickstein, and Samy Bengio. Density estimation using real nvp. arXiv:160508803.
68. Glow: generative flow with invertible 1x1 convolutions
69. Neural ordinary differential equations
70. Will Grathwohl, Ricky T. Q. Chen, Jesse Bettencourt, Ilya Sutskever, and David Duvenaud. FFJORD: free-form continuous dynamics for scalable reversible generative models. arXiv:181001367.
71. Training data augmentation: an empirical study using generative adversarial net-based approach with normalizing flow models for materials informatics10.1016/j.asoc.2019.105932 ↗
72. Language models are unsupervised multitask learners
73. Language models are few-shot learners
74. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. BERT: pre-training of deep bidirectional transformers for language understanding. arXiv:181004805.
75. Tong Xie, Yuwei Wan, Wei Huang, Yufei Zhou, Yixuan Liu, and Qingyuan Linghu et al. Large Language models as master key: unlocking the secrets of materials science with GPT. arXiv:230402213.
76. ChatGPT for Computational Materials Science: A Perspective10.34133/energymatadv.0026 ↗
77. Forecasting the outcome of spintronic experiments with neural ordinary differential equations10.1038/s41467-022-28571-7 ↗
78. Influences of separator thickness and surface coating on lithium dendrite growth: a phase-field study10.3390/ma15227912 ↗
79. Jason Wei, Yi Tay, Rishi Bommasani, Colin Raffel, Barret Zoph, and Sebastian Borgeaud et al. Emergent abilities of large language models. arXiv:220607682.
80. Machine learning embedded with materials domain knowledge
81. Deep generative models for materials discovery and machine learning-accelerated innovation10.3389/fmats.2022.865270 ↗
82. Data quantity governance for machine learning in materials science10.1093/nsr/nwad125 ↗
83. Detection method on data accuracy incorporating materials domain knowledge
84. Named entity recognition and normalization applied to large-scale information extraction from the materials science literature10.1021/acs.jcim.9b00470 ↗
85. Unsupervised word embeddings capture latent knowledge from materials science literature10.1038/s41586-019-1335-8 ↗
86. Automating materials exploration with a semantic knowledge graph for Li-ion battery cathodes10.1002/adfm.202201437 ↗
87. An automatic descriptors recognizer customized for materials science literature10.1016/j.jpowsour.2022.231946 ↗
88. Domain knowledge discovery from abstracts of scientific literature on Nickel-based single crystal superalloys
89. Heterogeneous catalyst design by generative adversarial network and first-principles based microkinetics10.1038/s41598-022-15586-9 ↗
90. Interfacegan: interpreting the disentangled face representation learned by GANs10.1109/TPAMI.2020.3034267 ↗
91. Cloud-VAE: variational autoencoder with concepts embedded10.1016/j.patcog.2023.109530 ↗
92. Laura von Rueden, Sebastian Mayer, Katharina Beckh, Bogdan Georgiev, Sven Giesselbach, and Raoul Heese et al. Informed machine learning--A taxonomy and survey of integrating knowledge into learning systems. arXiv:190312394.
93. High-throughput discovery of novel cubic crystal materials using deep generative neural networks10.1002/advs.202100566 ↗
94. A comprehensive survey of neural architecture search: challenges and solutions
95. SSRNAS: search space reduced one-shot NAS by a recursive attention-based predictor with cell tensor-flow diagram
96. Generative deep neural networks for inverse materials design using backpropagation and active learning
97. Jonathan Frankle, and Michael Carbin. The lottery ticket hypothesis: finding sparse, trainable neural networks. arXiv:180303635.
98. Andrew Brock, Jeff Donahue, and Karen Simonyan. Large scale GAN training for high fidelity natural image synthesis. arXiv:180911096.
99. Predictability and surprise in large generative models
100. Stealing machine learning models: attacks and countermeasures for generative adversarial networks