pith. sign in

arxiv: 2402.12365 · v5 · pith:XOE2ZGSEnew · submitted 2024-02-19 · 💻 cs.LG · cs.AI· physics.flu-dyn

Universal Physics Transformers: A Framework For Efficiently Scaling Neural Operators

classification 💻 cs.LG cs.AIphysics.flu-dyn
keywords simulationsuptsneuraloperatorsacrossdynamicslatentphysics
0
0 comments X
read the original abstract

Neural operators, serving as physics surrogate models, have recently gained increased interest. With ever increasing problem complexity, the natural question arises: what is an efficient way to scale neural operators to larger and more complex simulations - most importantly by taking into account different types of simulation datasets. This is of special interest since, akin to their numerical counterparts, different techniques are used across applications, even if the underlying dynamics of the systems are similar. Whereas the flexibility of transformers has enabled unified architectures across domains, neural operators mostly follow a problem specific design, where GNNs are commonly used for Lagrangian simulations and grid-based models predominate Eulerian simulations. We introduce Universal Physics Transformers (UPTs), an efficient and unified learning paradigm for a wide range of spatio-temporal problems. UPTs operate without grid- or particle-based latent structures, enabling flexibility and scalability across meshes and particles. UPTs efficiently propagate dynamics in the latent space, emphasized by inverse encoding and decoding techniques. Finally, UPTs allow for queries of the latent space representation at any point in space-time. We demonstrate diverse applicability and efficacy of UPTs in mesh-based fluid simulations, and steady-state Reynolds averaged Navier-Stokes simulations, and Lagrangian-based dynamics.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Accurate and scalable exchange-correlation with deep learning

    physics.chem-ph 2025-06 unverdicted novelty 7.0

    Skala is a neural XC functional trained on wavefunction data that beats state-of-the-art hybrids on main-group chemistry benchmarks at semi-local computational cost.

  2. Data-Efficient Neural Operator Training via Physics-Based Active Learning

    cs.LG 2026-05 unverdicted novelty 6.0

    Physics-based active learning using PDE residuals improves data efficiency for neural operator training on Burgers and Navier-Stokes equations while adding a physics inductive bias.

  3. Neuro-Symbolic AI for Analytical Solutions of Differential Equations

    cs.LG 2025-02 unverdicted novelty 6.0

    SIGS is a neuro-symbolic framework that discovers analytical solutions to PDEs by generating grammar-constrained expressions, embedding them in a topology-regularised latent manifold, and refining structure and coeffi...

  4. LEIA: Learned Environment for Interactive Architected Materials

    cs.LG 2026-05 unverdicted novelty 5.0

    LEIA is a world model for autoregressive 3D simulation of architected materials under interactive loading, benchmarked on MicroPlate and applied to surrogate-guided de novo design search with finite-element validation.

  5. Di-BiLPS: Denoising induced Bidirectional Latent-PDE-Solver under Sparse Observations

    cs.LG 2026-05 unverdicted novelty 5.0

    Di-BiLPS combines a variational autoencoder, latent diffusion, and contrastive learning to achieve state-of-the-art accuracy on PDE problems with as little as 3% observations while supporting zero-shot super-resolutio...