Active learning and automation will not easily liberate humans from laboratory workflows. Before they can really impact materials research, artificial intelligence systems will need to be carefully set up to ensure their robust operation and their ability to deal with both epistemic and stochastic errors. As autonomous experiments become more widely available, it is essential to think about how to embed reproducibility, reconfigurability and interoperability in the design of autonomous labs.
This is a preview of subscription content, access via your institution
Relevant articles
Open Access articles citing this article.
-
An active learning framework for the rapid assessment of galvanic corrosion
npj Materials Degradation Open Access 21 May 2024
-
A dynamic knowledge graph approach to distributed self-driving laboratories
Nature Communications Open Access 23 January 2024
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
$119.00 per year
only $9.92 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
References
Morgan, D. et al. Machine learning in nuclear materials research. Curr. Opin. Solid State Mater. Sci. 26, 100975 (2022).
Coley, C. W., Eyke, N. S. & Jensen, K. F. Autonomous discovery in the chemical sciences part II: outlook. Angew. Chem. Int. Ed. 59, 23414–23436 (2020).
Burger, B. et al. A mobile robotic chemist. Nature 583, 237–241 (2020).
Chen, J. et al. Navigating phase diagram complexity to guide robotic inorganic materials synthesis. Preprint at https://arxiv.org/abs/2304.00743 (2023).
Stach, E. et al. Autonomous experimentation systems for materials development: a community perspective. Matter 4, 2702–2726 (2021).
Siemenn, A. E., Ren, Z., Li, Q. & Buonassisi, T. Fast Bayesian optimization of needle-in-a-haystack problems using zooming memory-based initialization (ZoMBI). npj Comp. Mater. 9, 79 (2023).
Baker, M. 1,500 scientists lift the lid on reproducibility. Nature 533, 452–454 (2016).
Park, Y. J. et al. Can ChatGPT be used to generate scientific hypotheses? Preprint at https://arxiv.org/abs/2304.12208 (2023).
Arnold, C. Cloud labs: where robots do the research. Nature 606, 612–613 (2022).
Ren, Z. C., Zhang, Z., Tian Y. S. & Li, J. CRESt – Copilot for Real-world Experimental Scientist. Preprint at https://doi.org/10.26434/chemrxiv-2023-tnz1x (2023).
Acknowledgements
The authors thank Y. Tian for insightful discussions and R. S. Indradjaja for giving feedback on the manuscript. They acknowledge support by DTRA (award no. HDTRA1-20-2-0002) Interaction of Ionizing Radiation with Matter (IIRM) University Research Alliance (URA).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Competing interests
Zekun Ren and T.B. are co-founders of Xinterra Pte. Ltd, a startup focused on applying active learning to accelerate the development of materials for sustainability. The other authors declare no competing interests.
Rights and permissions
About this article
Cite this article
Ren, Z., Ren, Z., Zhang, Z. et al. Autonomous experiments using active learning and AI. Nat Rev Mater 8, 563–564 (2023). https://doi.org/10.1038/s41578-023-00588-4
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41578-023-00588-4
This article is cited by
-
A dynamic knowledge graph approach to distributed self-driving laboratories
Nature Communications (2024)
-
An active learning framework for the rapid assessment of galvanic corrosion
npj Materials Degradation (2024)