Design principles of deep translationally-symmetric neural quantum states for frustrated magnets
Rajah P. Nutakki, Ahmedeo Shokry, Filippo Vicentini · 2025
Read Design principles of deep translationally-symmetric neural quantum states for frustrated magnets by Rajah P. Nutakki, Ahmedeo Shokry, Filippo Vicentini (2025) 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/arxiv-2505-03466.txt.
Description
Rajah Nutakki, Ahmedeo Shokry, and Filippo Vicentini set design rules for neural network quantum states in frustrated magnets, in a 2025 paper. Such networks are a leading tool for finding the lowest energy state of quantum magnets. Frustrated magnets resist simple order, which makes them hard testbeds. Winning network designs lean on the magnet's shift symmetry, but why they win was unclear.
The team adapted a modern image network that blends convolution with transformer ideas. It proved strikingly close to a factored vision transformer already used on these magnets. That link ties the new design back to classic convolution networks. Careful trials tuned the design for best results at lowest cost. On two hard magnet models, its energies matched the best known, giving builders a clear blueprint.
Catalog record
| CREATOR | Rajah P. Nutakki, Ahmedeo Shokry, Filippo Vicentini |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 73 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-03466