An Efficient Graph-Transformer Operator for Learning Physical Dynamics with Manifolds Embedding

2025-12-10

AI Summary

Traditional numerical solvers incur prohibitive computational costs for problems involving complex geometries, dynamic boundaries, and multi-physics parameterizations; meanwhile, existing deep learning–based solvers—particularly those operating on unstructured meshes—suffer from poor generalization and limited flexibility. To address these limitations, we propose PhysGTO, a manifold-embedded Graph-Transformer operator. PhysGTO introduces a unified graph embedding module and a flux-guided message-passing mechanism that explicitly co-models the physical domain and its underlying implicit manifold structure. Its computational complexity scales linearly with mesh size, and it achieves substantial parameter compression while enabling real-time, high-fidelity, multi-scale physical simulation on unstructured meshes. Evaluated across 11 challenging benchmarks—including transient fluid flows, large-scale 3D geometries, and unstructured grids—PhysGTO attains state-of-the-art accuracy with significantly reduced FLOPs, outperforming prior methods in generalization, scalability, and robustness.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Sampling/Simulation-based Search
• Computer Vision: Low Level & Physics-based Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Accurate and efficient physical simulations are essential in science and engineering, yet traditional numerical solvers face significant challenges in computational cost when handling simulations across dynamic scenarios involving complex geometries, varying boundary/initial conditions, and diverse physical parameters. While deep learning offers promising alternatives, existing methods often struggle with flexibility and generalization, particularly on unstructured meshes, which significantly limits their practical applicability. To address these challenges, we propose PhysGTO, an efficient Graph-Transformer Operator for learning physical dynamics through explicit manifold embeddings in both physical and latent spaces. In the physical space, the proposed Unified Graph Embedding module aligns node-level conditions and constructs sparse yet structure-preserving graph connectivity to process heterogeneous inputs. In the latent space, PhysGTO integrates a lightweight flux-oriented message-passing scheme with projection-inspired attention to capture local and global dependencies, facilitating multilevel interactions among complex physical correlations. This design ensures linear complexity relative to the number of mesh points, reducing both the number of trainable parameters and computational costs in terms of floating-point operations (FLOPs), and thereby allowing efficient inference in real-time applications. We introduce a comprehensive benchmark spanning eleven datasets, covering problems with unstructured meshes, transient flow dynamics, and large-scale 3D geometries. PhysGTO consistently achieves state-of-the-art accuracy while significantly reducing computational costs, demonstrating superior flexibility, scalability, and generalization in a wide range of simulation tasks.

Problem

• Efficiently learning physical dynamics on unstructured meshes
• Reducing computational costs for complex geometry simulations
• Improving generalization across varying boundary conditions

Innovation

• Graph-Transformer Operator with manifold embeddings
• Unified Graph Embedding for heterogeneous inputs
• Lightweight flux-oriented message-passing with projection attention