Beyond Scalar Objectives: Expert-Feedback-Driven Autonomous Experimentation for Scientific Discovery at the Nanoscale.
8 affiliations
- 1Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee37831, United States.
- 2University of Tennessee-Oak Ridge Innovation Institute, University of Tennessee , Knoxville, Tennessee37996, United States.
- 3Department of Material Science and Engineering, School of Materials and Chemical Technology, Institute of Science Tokyo, Yokohama226-8502, Japan.
- 4School of Materials Science and Engineering, UNSW Sydney, Kensington, New South Wales2052, Australia.
- 5MacDiarmid Institute for Advanced Materials and Nanotechnology, School of Physical and Chemical Sciences, University of Canterbury, Christchurch8410, New Zealand.
- 6Department of Materials Science and Engineering, Norwegian University of Science and Technology (NTNU), TrondheimN-7491, Norway.
- 7Faculty of Physics and Center for Nanointegration Duisburg-Essen (CENIDE), University of Duisburg-Essen, Duisburg47057, Germany.
- 8Research Center Future Energy Materials and Systems, Research Alliance Ruhr, Bochum44780, Germany.
Abstract
Self-driving laboratories or autonomous experimentation are emerging as transformative platforms for accelerating scientific discovery. Bayesian optimization (BO) is among the most widely used machine learning frameworks in self-driving laboratories, but these BO-based frameworks rely on predefined scalar descriptors to guide experimentation. This limits their ability to handle complex phenomena that are difficult to formalize as scalar descriptors. In many situations, the determination of an appropriate scalar descriptor can be challenging, and may fail to capture subtle yet scientifically important phenomena apparent to experts with interdisciplinary insight. This problem is particularly pronounced in nanoscale imaging experiments, where strong local heterogeneity, multidimensional observations, and emergent physical behaviors often cannot be effectively represented using scalar metrics. To overcome this limitation, here we develop deep-kernel pairwise learning (DKPL), an approach for autonomous microscopy experiments which incorporates human expertise and interdisciplinary scientific knowledge into an active learning loop. Instead of relying on explicit scalar objectives, DKPL enables experts to directly evaluate which experimental output is more promising using interdisciplinary knowledge. DKPL then learns a latent utility function from these expert judgments to guide subsequent autonomous microscopy experiments. We demonstrate DKPL's performance in learning physically meaningful nanoscale structures in lead titanate thin films while effectively prioritizing high-information measurement regions using an experimental model dataset with known ground truth. We further apply DKPL to analyze the character of ferroelectric domain walls, where we find DKPL capable of distinguishing between high and low characteristic domain-wall angles in lead zirconium titanate, and able to discover both head-to-head and tail-to-tail domain-wall character in erbium manganite. Autonomous experiments in mixed-phase bismuth ferrite also demonstrate DKPL's ability to link material structure with their properties in on-the-fly measurements. This development demonstrates an approach to integrate expert knowledge into autonomous microscopy experiments across a variety of material systems, and demonstrates a pathway toward expert-guided self-driving laboratories capable of addressing scientific problems beyond the limits of scalar-metrics-driven learning.
Related
- LLM-AutoSciLab: Closed-Loop Scientific Discovery via Active Experimentation with LLMs
- Scalable Thermoset Informatics Integrated with Autonomous Experimentation for On-Demand Discovery
- Autonomous in-silico inorganic materials discovery via multi-agent physics-aware scientific reasoning
- Autonomous Experimentation Enables Rapid Optimization of Photochemical Continuous-Flow Synthesis of Diazaxanthilidene Fluorophores
- Agentic AI for Scientific Reasoning in Autonomous Quantum Sensing Experiments
BibTeX
@article{bulanadi2026beyond,
title = {{Beyond Scalar Objectives: Expert-Feedback-Driven Autonomous Experimentation for Scientific Discovery at the Nanoscale.}},
author = {Ralph Bulanadi and Jefferey Baxter and Arpan Biswas and Hiroshi Funakubo and Valanoor Nagarajan and Daniel Sando and Dennis Meier and Jan Schultheiß and Rama Vasudevan and Yongtao Liu},
journal = {ACS nano},
year = {2026},
month = oct,
doi = {10.1021/acsnano.6c09449},
url = {https://europepmc.org/article/MED/42899772},
note = {CC BY},
}
RIS
TY - JOUR
TI - Beyond Scalar Objectives: Expert-Feedback-Driven Autonomous Experimentation for Scientific Discovery at the Nanoscale.
AU - Ralph Bulanadi
AU - Jefferey Baxter
AU - Arpan Biswas
AU - Hiroshi Funakubo
AU - Valanoor Nagarajan
AU - Daniel Sando
AU - Dennis Meier
AU - Jan Schultheiß
AU - Rama Vasudevan
AU - Yongtao Liu
PY - 2026
DA - 2026/10/01
JO - ACS nano
DO - 10.1021/acsnano.6c09449
UR - https://europepmc.org/article/MED/42899772
AB - Self-driving laboratories or autonomous experimentation are emerging as transformative platforms for accelerating scientific discovery. Bayesian optimization (BO) is among the most widely used machine learning frameworks in self-driving laboratories, but these BO-based frameworks rely on predefined scalar descriptors to guide experimentation. This limits their ability to handle complex phenomena that are difficult to formalize as scalar descriptors. In many situations, the determination of an appropriate scalar descriptor can be challenging, and may fail to capture subtle yet scientifically important phenomena apparent to experts with interdisciplinary insight. This problem is particularly pronounced in nanoscale imaging experiments, where strong local heterogeneity, multidimensional observations, and emergent physical behaviors often cannot be effectively represented using scalar metrics. To overcome this limitation, here we develop deep-kernel pairwise learning (DKPL), an approach for autonomous microscopy experiments which incorporates human expertise and interdisciplinary scientific knowledge into an active learning loop. Instead of relying on explicit scalar objectives, DKPL enables experts to directly evaluate which experimental output is more promising using interdisciplinary knowledge. DKPL then learns a latent utility function from these expert judgments to guide subsequent autonomous microscopy experiments. We demonstrate DKPL's performance in learning physically meaningful nanoscale structures in lead titanate thin films while effectively prioritizing high-information measurement regions using an experimental model dataset with known ground truth. We further apply DKPL to analyze the character of ferroelectric domain walls, where we find DKPL capable of distinguishing between high and low characteristic domain-wall angles in lead zirconium titanate, and able to discover both head-to-head and tail-to-tail domain-wall character in erbium manganite. Autonomous experiments in mixed-phase bismuth ferrite also demonstrate DKPL's ability to link material structure with their properties in on-the-fly measurements. This development demonstrates an approach to integrate expert knowledge into autonomous microscopy experiments across a variety of material systems, and demonstrates a pathway toward expert-guided self-driving laboratories capable of addressing scientific problems beyond the limits of scalar-metrics-driven learning.
SN - 1936-0851
SN - 1936-086X
ER -