DISCOVER: A Physics-Informed, GPU-Accelerated Symbolic Regression Framework
Udaykumar Gajera, Mohsen Sotoudeh, Kanchan Sarkar, Axel Groß · 2026
Read DISCOVER: A Physics-Informed, GPU-Accelerated Symbolic Regression Framework by Udaykumar Gajera, Mohsen Sotoudeh, Kanchan Sarkar, Axel Groß (2026) 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-2602-06986.txt.
Description
Udaykumar Gajera and three co-authors present DISCOVER, an open-source package for finding clear math relationships in physics, chemistry, and materials data. Such relationships, often called descriptors, link a basic material property to a wanted function. Known tools have found low-dimensional descriptors, but many fit poorly with Python workflows, give limited control over the search space, or strain at large scale.
DISCOVER answers those gaps with a modular, physics-minded design: users steer the search with domain knowledge, fence the feature space plainly, and use optional GPU speed-up for heavy studies. The package stresses physically real models, same-result-twice reruns, and scale-up, complementing general-purpose tools. It targets settings where clear models, physical agreement, and run time matter most.
Catalog record
| CREATOR | Udaykumar Gajera, Mohsen Sotoudeh, Kanchan Sarkar, Axel Groß |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 11 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-06986