journal article·2026-01-14

Pareto-Optimal Experimentation: Human-Guided Multi-Objective Bayesian Optimization in Scanning Probe Microscopy

Yu Liu, Sergei V. Kalinin

Nano Letters· Volume 26· Issue 1
Publisher page

Abstract

Automated experimentation has the potential to accelerate scientific discovery across disciplines, but its success requires systematic methods to optimize multiple, often conflicting, objectives under uncertainty. We present multi-objective Bayesian optimization (MOBO) as a general framework for balancing competing rewards and integration of human guidance in autonomous experimentation. Rather than identifying a single optimum, MOBO constructs the Pareto front, providing a principled description of all trade-off solutions and revealing the interdependencies between partially known reward functions. This enables systematic exploration of parameter space and quantifiable decision-making. Importantly, MOBO naturally supports human-in-the-loop control. Researchers can reweight objectives or adjust reference points to steer experiments toward desired outcomes, incorporating expert judgment without breaking automation. Together, MOBO and human guidance transform experimental optimization from trial-and-error tuning into a reproducible, interpretable, and customizable process, offering a scalable methodology for building trustworthy and efficient self-driving laboratories.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Force Microscopy Techniques and Applications · Model Reduction and Neural Networks · Piezoelectric Actuators and Control

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

Affiliations

Department of Materials Science and Engineering

University of Tennessee

Physical Sciences Division

Pacific Northwest National Laboratory

Funding

National Science Foundation

References (51)

Show source-provided references
1. 10.1038/s41586-023-06221-210.1038/s41586-023-06221-2 ↗
2. 10.1038/s41540-021-00189-310.1038/s41540-021-00189-3 ↗
3. 10.1021/acsnano.4c1876010.1021/acsnano.4c18760 ↗
4. 10.1021/acsnano.1c1111810.1021/acsnano.1c11118 ↗
5. 10.1063/5.024968610.1063/5.0249686 ↗
6. 10.1039/D4DD00115J10.1039/D4DD00115J ↗
7. 10.1021/acsnano.4c1750410.1021/acsnano.4c17504 ↗
8. 10.1021/acs.chemrev.4c0005510.1021/acs.chemrev.4c00055 ↗
9. 10.1039/D4DD00040D10.1039/D4DD00040D ↗
10. 10.1038/s44286-023-00002-410.1038/s44286-023-00002-4 ↗
11. 10.1021/acssuschemeng.4c0217710.1021/acssuschemeng.4c02177 ↗
12. 10.1038/s41467-023-44599-910.1038/s41467-023-44599-9 ↗
13. 10.1021/acs.chemmater.2c0359310.1021/acs.chemmater.2c03593 ↗
14. 10.1038/s44160-022-00231-010.1038/s44160-022-00231-0 ↗
15. 10.1038/s41467-022-28580-610.1038/s41467-022-28580-6 ↗
16. 10.1021/acs.accounts.2c0022010.1021/acs.accounts.2c00220 ↗
17. 10.1126/sciadv.aaz886710.1126/sciadv.aaz8867 ↗
18. 10.1063/5.016996110.1063/5.0169961 ↗
19. 10.1002/adma.20220134510.1002/adma.202201345 ↗
20. 10.1038/s41524-020-00487-010.1038/s41524-020-00487-0 ↗
21. 10.1002/adma.20210368010.1002/adma.202103680 ↗
22. 10.1021/jacs.2c0508210.1021/jacs.2c05082 ↗
23. 10.1038/s41578-021-00337-510.1038/s41578-021-00337-5 ↗
24. 10.48550/arXiv.2412.1806710.48550/arXiv.2412.18067 ↗
25. 10.1126/sciadv.adu742610.1126/sciadv.adu7426 ↗
26. 10.1039/D5DD00033E10.1039/D5DD00033E ↗
27. 10.1038/s41524-023-00985-x10.1038/s41524-023-00985-x ↗
28. 10.1038/s42256-022-00460-010.1038/s42256-022-00460-0 ↗
29. 10.1039/D4DD00132J10.1039/D4DD00132J ↗
30. 10.1002/adma.20241892710.1002/adma.202418927 ↗
31. 10.1002/advs.20240244010.1002/advs.202402440 ↗
32. 10.1002/advs.20241064010.1002/advs.202410640 ↗
33. 10.1145/361053610.1145/3610536 ↗
34. 10.1007/s10462-022-10359-210.1007/s10462-022-10359-2 ↗
35. 10.1109/SSCI.2016.785022110.1109/SSCI.2016.7850221 ↗
36. 10.3390/computation1003003710.3390/computation10030037 ↗
37. 10.1038/s43586-021-00033-210.1038/s43586-021-00033-2 ↗
38. 10.1063/1.11179510.1063/1.111795 ↗
39. 10.1126/science.292879410.1126/science.2928794 ↗
40. 10.1126/science.260.5113.145110.1126/science.260.5113.1451 ↗
41. 10.1103/PhysRevLett.56.93010.1103/PhysRevLett.56.930 ↗
42. 10.1103/PhysRevLett.49.5710.1103/PhysRevLett.49.57 ↗
43. 10.1109/CCA.2009.528104610.1109/CCA.2009.5281046 ↗
44. 10.1109/TCST.2005.84733110.1109/TCST.2005.847331 ↗
45. 10.1063/1.10469010.1063/1.104690 ↗
46. 10.1088/0957-4484/21/45/45570510.1088/0957-4484/21/45/455705 ↗
47. 10.1115/1.400015810.1115/1.4000158 ↗
48. 10.48550/arXiv.1910.0640310.48550/arXiv.1910.06403 ↗
49. 10.48550/arXiv.2105.0819510.48550/arXiv.2105.08195 ↗
50. 10.48550/arXiv.1809.1116510.48550/arXiv.1809.11165 ↗
51. 10.1063/5.021999010.1063/5.0219990 ↗