2 Comments
User's avatar
Piyush's avatar

super cool! I am also working on material discovery using these models. it looks close but just out of reach

chemicallyGeeky's avatar

Great article! The part on the difficulties of trusting the literature really resonate. I wrote a piece on this topic here: Materials Science Has a Reproducibility Problem. Machine Learning Inherits It.

https://chemicallygeeky.substack.com/p/materials-science-has-a-reproducibility?r=4and5k