Spectral operator representations
Austin Zadoks, Antimo Marrazzo, Nicola Marzari · 2024
Read Spectral operator representations by Austin Zadoks, Antimo Marrazzo, Nicola Marzari (2024) 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-11611740.txt.
Description
Austin Zadoks, Antimo Marrazzo, and Nicola Marzari set out, in a 2024 paper, a new way to feed materials data to machine learning. Most tools read a material by its atom layout, split into local zones. That suits force models well. But traits such as band gaps and charge flow live in a wave view of the material, not in neat local zones. For those traits, a form that reads the electron layout may work better. Their frame builds such forms from spectral operators, which keep the math's natural symmetry and sense.
They test the frame first on how alike two materials are. They then use it to screen for clear conductors, a class prized for screens and solar cells. Trained on 217 materials, the model flagged three of four true hits in a large 3D database. The work widens the toolkit for fast finds.
Catalog record
| CREATOR | Austin Zadoks, Antimo Marrazzo, Nicola Marzari |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |