Interpretable Bayesian optimization for catalyst discovery
Akhil S. Nair, Lucas Foppa, Matthias Scheffler · 2026
Read Interpretable Bayesian optimization for catalyst discovery by Akhil S. Nair, Lucas Foppa, Matthias Scheffler (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13127247.txt.
Description
Akhil Nair, Lucas Foppa, and Matthias Scheffler present a clearer way to search for new catalysts in this 2026 paper. Catalysts speed up chemical reactions, and computer-guided search can test many candidate materials. Standard search tools need a few key physical traits in advance, but for these reactions the right traits are unknown.
Their framework picks the key traits as it searches, using a method that builds short formulas from physical quantities. It can catch curved links and interactions between traits that simpler screens miss. In simulated hunts, the method found surface sites of one- and two-atom alloys that can activate carbon dioxide. It beat widely used feature-picking approaches. The authors aim to make guided discovery work where expert traits are not known ahead.
Catalog record
| CREATOR | Akhil S. Nair, Lucas Foppa, Matthias Scheffler |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |