journal article·2025-08-19

Navigating Small Datasets with Machine Learning: Gaussian Process Modeling for Colloidal Gelation

Rohan Batra, Yogesh M. Joshi, Sachin Shanbhag

Langmuir· Volume 41· Issue 32
Publisher page

Abstract

Small datasets, which are common in many traditional science and engineering disciplines, present a challenge for powerful machine learning methods that are better suited for large datasets. Through the example of spontaneous colloidal sol-gel transition, we present a data-driven framework based on Gaussian Processes (GPs) for modeling and inference with limited data. GPs are well-suited for small datasets and automatically provide uncertainty quantification for model predictions. We explore GPs with two different structures (single and multi-output) to map synthesis parameters to gelation characteristics. While both GP structures quantitatively capture the relationships between the input and output variables, we favor multioutput GP, which attempts to simultaneously learn correlations between different outputs, due to its superior uncertainty calibration. To address inverse problems, we introduce a Bayesian framework that embeds a GP model into a Markov Chain Monte Carlo sampling algorithm. It enables the discovery of multiple synthesis pathways for achieving desired material properties. Therefore, this work presents a robust methodology for data-constrained modeling, with broad applicability to material design in general and the colloidal domain in particular.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Gaussian Processes and Bayesian Inference · Machine Learning in Materials Science · Phase Equilibria and Thermodynamics

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

Affiliations

Department of Chemical Engineering

Indian Institute of Technology Kanpur

Department of Scientific Computing

Florida State University

Funding

Science and Engineering Research Board · Florida State University

References (75)

Show source-provided references
1. Bishop, C. M. Pattern recognition and machine learning, Springer-Verlag: New York, 2006; p. 778.
2. Brunton S. L. · 202210.1017/9781009089517 ↗
3. 10.1021/jp500350b10.1021/jp500350b ↗
4. 10.1021/acs.langmuir.2c0335510.1021/acs.langmuir.2c03355 ↗
5. 10.1016/j.jcis.2022.06.03410.1016/j.jcis.2022.06.034 ↗
6. 10.1021/acs.langmuir.4c0067410.1021/acs.langmuir.4c00674 ↗
7. 10.1016/j.jcis.2021.11.19510.1016/j.jcis.2021.11.195 ↗
8. 10.1039/D1SM01466H10.1039/D1SM01466H ↗
9. 10.1016/j.colcom.2022.10059510.1016/j.colcom.2022.100595 ↗
10. 10.1007/s00397-023-01407-x10.1007/s00397-023-01407-x ↗
11. 10.1016/j.cocis.2024.10187310.1016/j.cocis.2024.101873 ↗
12. 10.1021/acs.langmuir.3c0383110.1021/acs.langmuir.3c03831 ↗
13. 10.1021/acs.langmuir.4c0355310.1021/acs.langmuir.4c03553 ↗
14. 10.1021/acs.langmuir.3c0187810.1021/acs.langmuir.3c01878 ↗
15. 10.1021/acs.langmuir.4c0408810.1021/acs.langmuir.4c04088 ↗
16. 10.1021/acs.langmuir.4c0105010.1021/acs.langmuir.4c01050 ↗
17. 10.1021/acs.langmuir.3c0268810.1021/acs.langmuir.3c02688 ↗
18. 10.1017/dce.2023.310.1017/dce.2023.3 ↗
19. 10.1007/s00397-023-01405-z10.1007/s00397-023-01405-z ↗
20. 10.1073/pnas.230466912010.1073/pnas.2304669120 ↗
21. 10.1007/s00397-023-01416-w10.1007/s00397-023-01416-w ↗
22. 10.1103/PhysRevResearch.2.03310710.1103/PhysRevResearch.2.033107 ↗
23. 10.1122/8.000087210.1122/8.0000872 ↗
24. Cawley G. C. · J. Mach. Learn. Res. · 2010 · 11 · 2079
25. 10.3748/wjg.v28.i5.60510.3748/wjg.v28.i5.605 ↗
26. 10.1007/BF0193228510.1007/BF01932285 ↗
27. Tikhonov A. I. N. · 199510.1007/978-94-015-8480-7 ↗
28. 10.1002/mats.20190000510.1002/mats.201900005 ↗
29. 10.1007/s00397-020-01212-w10.1007/s00397-020-01212-w ↗
30. Adcock B. · 202210.1137/1.9781611976885 ↗
31. 10.1016/j.jcp.2018.10.04510.1016/j.jcp.2018.10.045 ↗
32. 10.1017/jfm.2016.61510.1017/jfm.2016.615 ↗
33. 10.1063/5.023360710.1063/5.0233607 ↗
34. Rasmussen C. E. · 2006
35. Matheron, G. The theory of regionalized variables and its applications, Ecole des Mines de Paris: Paris, 1971; p. 212.
36. Hensman J. · arXiv · 2013
37. 10.1142/S021987622143003910.1142/S0219876221430039 ↗
38. 10.1016/j.engappai.2023.10610710.1016/j.engappai.2023.106107 ↗
39. 10.1007/s10973-020-09990-410.1007/s10973-020-09990-4 ↗
40. 10.1038/s41598-020-74394-110.1038/s41598-020-74394-1 ↗
41. 10.1063/5.004738510.1063/5.0047385 ↗
42. 10.1557/s43577-021-00183-410.1557/s43577-021-00183-4 ↗
43. Snoek J. · arXiv · 2012
44. Frazier, P. I.; Wang, J. Information Science for Materials Discovery and Design, Lookman, T.; Alexander, F. J.; Rajan, K.; Eds. Springer International Publishing, 2016; pp. 45–75.
45. 10.1039/D1SM00250C10.1039/D1SM00250C ↗
46. 10.1016/j.cartre.2023.10025210.1016/j.cartre.2023.100252 ↗
47. 10.1016/j.colsurfa.2024.13605710.1016/j.colsurfa.2024.136057 ↗
48. 10.1146/annurev-fluid-031821-10493510.1146/annurev-fluid-031821-104935 ↗
49. 10.1122/1.513411510.1122/1.5134115 ↗
50. Winter, H. H.; Mours, M. Neutron Spin Echo Spectroscopy Viscoelasticity Rheology, 1st ed. Springer: Berlin Heidelberg, 1997; pp. 165–234.
51. 10.1063/5.003883010.1063/5.0038830 ↗
52. 10.1021/acs.langmuir.7b0015110.1021/acs.langmuir.7b00151 ↗
53. 10.1021/acs.langmuir.8b0183010.1021/acs.langmuir.8b01830 ↗
54. 10.1122/8.000074310.1122/8.0000743 ↗
55. 10.1122/1.352362710.1122/1.3523627 ↗
56. 10.1021/ma200477210.1021/ma2004772 ↗
57. Gardner J. R. · arXiv · 2018
58. Paszke A. · 2019
59. Bonilla E. V. · Adv. Neural Inf. Process. Syst. · 2007 · 20
60. 10.1109/TNNLS.2019.295710910.1109/TNNLS.2019.2957109 ↗
61. Williams C. · Adv. Neural Inf. Process. Syst. · 2000 · 13
62. Snelson E. · 2006 · 1257
63. Snelson E. · 2007 · 2 · 524
64. 10.1137/17M111294210.1137/17M1112942 ↗
65. 10.1007/s11222-019-09886-w10.1007/s11222-019-09886-w ↗
66. 10.1007/s11222-022-10124-z10.1007/s11222-022-10124-z ↗
67. Melkumyan A. · 2009 · 1936
68. Rasmussen C. · Adv. Neural Inf. Process. Syst. · 2001 · 14
69. Park C. · J. Mach. Learn. Res. · 2016 · 17 · 6071
70. Park C. · J. Mach. Learn. Res. · 2018 · 19 · 1
71. Deisenroth M. · arXiv · 2015
72. 10.1016/j.engappai.2021.10444910.1016/j.engappai.2021.104449 ↗
73. 10.1063/1.169911410.1063/1.1699114 ↗
74. 10.1021/ma00101a02810.1021/ma00101a028 ↗
75. 10.1038/s41567-023-01988-710.1038/s41567-023-01988-7 ↗