Dataset Design for Building Models of Chemical Reactivity

Priyanka Raghavan, Brittany C. Haas, Madeline E. Ruos, Jules Schleinitz, Abigail G. Doyle, Sarah E. Reisman, Matthew S. Sigman, Connor W. Coley · 2023

Read Dataset Design for Building Models of Chemical Reactivity by Priyanka Raghavan, Brittany C. Haas, Madeline E. Ruos, Jules Schleinitz, Abigail G. Doyle, Sarah E. Reisman, Matthew S. Sigman, Connor W. Coley (2023) 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-10755851.txt.

Description

Priyanka Raghavan and co-authors argue that dataset design decides whether models of chemical reactivity succeed, in a 2023 perspective paper. Models that predict reaction outcomes are only as good as their training data, and a model can only cover the territory its data actually spans.

The authors explain how the choice of molecular or reaction representation shapes both diversity and reach. They also lay out the practical limits of common chemistry datasets, since bench experiments are slow and selective. Their guidance helps chemists plan data collection with the final model in mind. Better-designed sets, they contend, are the shortest path to trustworthy predictions.

Catalog record

CREATORPriyanka Raghavan, Brittany C. Haas, Madeline E. Ruos, Jules Schleinitz, Abigail G. Doyle, Sarah E. Reisman, Matthew S. Sigman, Connor W. Coley
YEAR2023
COLLECTIONPAPERS
FORMATTXT
FILE SIZE30 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC10755851

Browse the full library · About the Tower of Babel