Metraipolis › archive › dc_5819317385ea

Physics-Informed Neural Plasticity: PDE Solvers That Reshape Themselves — arXiv:2610.09510

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09510
License (as archived)
CC0-1.0
Archived
2026-10-08T11:56:41.427Z
SHA-256 of the archived text
c4b86a1c7facb012b0f385a7bec336e398b8bdad0a6cf1fa33fb784ca6466fee
Stable URL of this page
https://metraipolis.com/documento/dc_5819317385ea

How to cite

Physics-Informed Neural Plasticity: PDE Solvers That Reshape Themselves — arXiv:2610.09510 — arXiv (en), https://arxiv.org/abs/2610.09510 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5819317385ea (Metraipolis archive dc_5819317385ea, sha256 c4b86a1c7facb012b0f385a7bec336e398b8bdad0a6cf1fa33fb784ca6466fee)
BibTeX
@misc{dc5819317385ea,
  title = {Physics-Informed Neural Plasticity: PDE Solvers That Reshape Themselves — arXiv:2610.09510},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09510},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5819317385ea; sha256 c4b86a1c7facb012b0f385a7bec336e398b8bdad0a6cf1fa33fb784ca6466fee},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09510. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Physics-informed neural PDE solvers adapt their parameters to satisfy governing equations, yet their representational structure typically remains fixed throughout training. This rigidity is poorly matched to PDE solutions with strongly heterogeneous complexity across space and space--time, leaving capacity insufficient where the physics is difficult and redundant where it is simple. We introduce physics-informed neural plasticity, a paradigm in which the representation itself reshapes during optimization in response to unresolved physics. We instantiate this principle with Representation Capacity Adaptation for PDEs (ReCAP), a Gaussian-localized solver that dynamically redistributes capacity through local enrichment, residual-directed splitting, gate-based pruning, and function-aware merging. ReCAP uses responsibility-weighted error indicators and the geometry of residual energy to determine where and how to refine. To limit the disturbance introduced by splitting, we introduce quiet-child refinement, which initializes new components by transporting the parent representation while controlling instantaneous functional perturbation. We further establish conditional a posteriori reliability and structural-stability guarantees linking localized physics residuals to solution error and stable refinement. Across five challenging 3D and 4D PDE benchmarks against 11 physics-informed solvers, ReCAP achieves the lowest relative $L^2$ error on every problem, reducing error by $10.7\%$--$27.5\%$ relative to the strongest competing result. These results suggest that physics-informed solvers need not merely learn their parameters---they can learn how their representational capacity should be organized.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Numerical Analysis (math.NA).
Pagina del lavoro: https://arxiv.org/abs/2610.09510 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T10:50:37.689Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).