A Physics-preserved Transfer Learning Method for Differential Equations
Hao-Ran Yang, Chuan-Xian Ren · 2025
Read A Physics-preserved Transfer Learning Method for Differential Equations by Hao-Ran Yang, Chuan-Xian Ren (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-2505-01281.txt.
Description
Hao-Ran Yang and Chuan-Xian Ren give a transfer method that moves a learned solver for change rules to a new site while it keeps the physics true, in a paper from 2025. Learned solvers shine on their home data but sag when the site shifts, by bent data or by a shift in the rule itself. Past transfer tricks lack reach across rule types or drop the physics on the way.
The method reads a data site as a joint spread and splits the shift into data bias and rule bias. A haul map built with physics care then warps the home spread to the target. Wide tests show top scores, broad reach across rule types, and true physics held through the move. The authors pitch it as a shared tool for field shifts.
Catalog record
| CREATOR | Hao-Ran Yang, Chuan-Xian Ren |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 57 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-01281