Quasi-van Hove singularities informed approach improving DOS/pDOS predictions in GNN
Grigory Koroteev · 2025
Read Quasi-van Hove singularities informed approach improving DOS/pDOS predictions in GNN by Grigory Koroteev (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-2509-02818.txt.
Description
Grigory Koroteev wrote a 2025 solo materials informatics paper on improving machine learning predictions of the density of states, the count of electron energy levels available in a material. Graph neural networks predict that density, and its projected variants, but sharp features are often missed.
Koroteev extends the idea of van Hove singularities, points where the density of states spikes, into a quasi form usable as prior information for learning models. Instead of computing the density exactly, the method works with model estimates and adds a post-processing step guided by that extra information. Regularisation, described as close in spirit to Fisher information weighting, redistributes what the model knows about the system. The author reports significantly better prediction quality, with the gain likened to higher-order singularity effects.
Catalog record
| CREATOR | Grigory Koroteev |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 25 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-02818