Skip to content
AI News HubLIVE
Original source2 min read

Separable Neural Architectures as Physical World Models: from Mathematical Theory to Applications

Summary

The paper introduces the Separable Neural Architecture (SNA), combining neural approximation with tensor decomposition to solve high-dimensional PDEs under a variational framework, and demonstrates it as a 'solve once, query anywhere' physical world model with large speedups and real-time inversion in engineering case studies.

SourcearXiv Machine LearningAuthor: Reza T Batley, Andrew Kichline, Sourav Saha
Separable Neural Architectures as Physical World Models: from Mathematical Theory to Applications
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[2606.14934] Separable Neural Architectures as Physical World Models: from Mathematical Theory to Applications

[Submitted on 12 Jun 2026]

Title:Separable Neural Architectures as Physical World Models: from Mathematical Theory to Applications

View a PDF of the paper titled Separable Neural Architectures as Physical World Models: from Mathematical Theory to Applications, by Reza T Batley and Andrew Kichline and Sourav Saha

View PDF HTML (experimental)

Abstract:This work introduces the Separable Neural Architecture (SNA), a function representational class combining neural approximation with tensor decomposition. The SNA decouples localized coordinate functions (atoms) from global interactions governed by a sparse, low-rank interaction object. This architecture possesses a compact and smooth inductive bias well-suited for solving partial differential equations (PDEs). When viewed as a Galerkin trial space under the variational SNA (VSNA) framework, the formulation satisfies classical variational guarantees under Lax-Milgram: well-posedness, quasi-optimality, convergence, and stability. In high-dimensional spatiotemporal--parametric PDEs, the VSNA mitigates the curse of dimensionality by scaling algebraically rather than exponentially. Exploiting an entirely factorized, tensor-native alternating least squares (ALS) optimization framework reduces this cost to linear in dimension. The VSNA is validated across elliptic, hyperbolic, and parabolic systems, demonstrating close alignment with predicted algebraic and spectral scaling rates. We showcase the SNA as a "solve once, query anywhere" physical world model via two engineering case studies: a 7D parametric manufacturing simulation and an experimental thermal-to-property inversion pipeline for Inconel 718. The VSNA executes a 1,000,000-query Monte Carlo sweep in 102s on a standard laptop CPU, yielding a 150,000x speedup over a full-grid finite element baseline hosted on an NVIDIA A100 GPU. It further enables real-time generative inverse-mode reconstructions under 100ms. These results demonstrate that the SNA serves as a compact mathematical substrate for continuous parameter manifolds to enable real-time inversion, optimization loops, and rapid uncertainty propagation.

Subjects:

Machine Learning (cs.LG); Artificial Intelligence (cs.AI)

Cite as: arXiv:2606.14934 [cs.LG]

(or arXiv:2606.14934v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2606.14934

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Reza Batley [view email] [v1] Fri, 12 Jun 2026 20:14:49 UTC (4,210 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Separable Neural Architectures as Physical World Models: from Mathematical Theory to Applications, by Reza T Batley and Andrew Kichline and Sourav Saha

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.LG

new | recent | 2026-06

Change to browse by:

cs cs.AI

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

IArxiv recommender toggle

IArxiv Recommender (What is IArxiv?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • SNA decouples localized coordinate functions from global sparse, low-rank interactions, yielding a compact inductive bias suited for PDEs.
  • Under the VSNA framework, it satisfies Lax-Milgram variational guarantees and scales linearly with dimension, mitigating the curse of dimensionality.
  • In a 7D manufacturing simulation and Inconel 718 thermal-to-property inversion, a 1,000,000-query Monte Carlo sweep runs in 102s, a 150,000x speedup over a full-grid FEM baseline on an A100 GPU.
  • It enables real-time generative inverse-mode reconstructions under 100ms, supporting optimization loops and uncertainty propagation.

Highlights and analysis are generated automatically and may contain errors. Check the original source.