APIC: Orthogonalized Neuro-Symbolic Modeling for Nonlinear Dissipative Dynamics
Yanhui Zhu, Xiangfu Meng, Chen Zhao, Yinhao Li
Abstract
Current data-driven scientific modeling struggles with a functional dichotomy: neural operators exhibit spectral bias in high-frequency regimes, while physics-constrained paradigms suffer from optimization pathologies. To bridge this gap, we propose Adaptive Physics-Informed Computing (APIC), a neuro-symbolic meta-architecture designed with structural reconfigurability to encode diverse domain priors. Crucially, APIC integrates a gradient isolation strategy that reduces interference between the optimization paths of parameter identification and residual correction, effectively mitigating gradient conflicts. By instantiating this framework for nonlinear dissipative systems, we derive the Generalized Kuramoto-Sivashinsky-Cahn-Hilliard (G-KSCH) kernel, providing a unified representation for sparse dynamic identification. Extensive experiments demonstrate that APIC establishes new benchmarks in 3D compressible supersonic shock wave prediction, surpassing diverse architectures (e.g., CNNs and Transformers) by substantial margins in predictive accuracy. Notably, APIC achieves Pareto-optimal performance, delivering superior precision with reduced computational overhead compared to SOTA models, while exhibiting strong cross-task adaptability across meteorological and urban traffic datasets.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on6
- Fourier Neural Operator for Parametric Partial Differential EquationsZongyi Li, Nikola Borislavov Kovachki, Kamyar Azizzadenesheli, Burigede Liu et al.ICLR 2021 · 3,911 citations
- Solver-in-the-Loop: Learning from Differentiable Physics to Interact with Iterative PDE-SolversKiwon Um, Robert Brand, Yun (Raymond) Fei, Philipp Holl et al.NeurIPS 2020 · 398 citations
- GNOT: A General Neural Operator Transformer for Operator LearningZhongkai Hao, Zhengyi Wang, Hang Su, Chengyang Ying et al.ICML 2023 · 375 citations
- Learning to Control PDEs with Differentiable PhysicsPhilipp Holl, Nils Thuerey, Vladlen KoltunICLR 2020 · 221 citations
- Disentangling Physical Dynamics From Unknown Factors for Unsupervised Video PredictionVincent Le Guen, Nicolas ThomeCVPR 2020
Related papers
- Bias-Spectrum Neural Processes for Parametric PDEs: Architecture Priors Meet PDE ConstraintsHui Li, Huafeng Liu, Chenguang Li, Tianxiao Zhang et al.ICML 2026
- PARCv2: Physics-aware Recurrent Convolutional Neural Networks for Spatiotemporal Dynamics ModelingPhong C. H. Nguyen, Xinlun Cheng, Shahab Azarfar, Pradeep K. Seshadri et al.ICML 2024 · 12 citations
- Error-Driven Graph Augmentation for Mesh-Based PDE SurrogatesXuan Minh Vuong Nguyen, Nissrine Akkari, Fabien Casenave, Jonathan Viquerat et al.ICML 2026
- MaNGO - Adaptable Graph Network Simulators via Meta-LearningPhilipp Dahlinger, Tai Hoang, Denis Blessing, Niklas Freymuth et al.NeurIPS 2025 · 4 citations
- Adaptive Mamba Neural OperatorsZeyuan Song, Zheyu JiangICLR 2026 · 6 citations
