-
KP-PINNs: Kernel Packet Accelerated Physics Informed Neural Networks
Authors:
Siyuan Yang,
Cheng Song,
Zhilu Lai,
Wenjia Wang
Abstract:
Differential equations are involved in modeling many engineering problems. Many efforts have been devoted to solving differential equations. Due to the flexibility of neural networks, Physics Informed Neural Networks (PINNs) have recently been proposed to solve complex differential equations and have demonstrated superior performance in many applications. While the L2 loss function is usually a de…
▽ More
Differential equations are involved in modeling many engineering problems. Many efforts have been devoted to solving differential equations. Due to the flexibility of neural networks, Physics Informed Neural Networks (PINNs) have recently been proposed to solve complex differential equations and have demonstrated superior performance in many applications. While the L2 loss function is usually a default choice in PINNs, it has been shown that the corresponding numerical solution is incorrect and unstable for some complex equations. In this work, we propose a new PINNs framework named Kernel Packet accelerated PINNs (KP-PINNs), which gives a new expression of the loss function using the reproducing kernel Hilbert space (RKHS) norm and uses the Kernel Packet (KP) method to accelerate the computation. Theoretical results show that KP-PINNs can be stable across various differential equations. Numerical experiments illustrate that KP-PINNs can solve differential equations effectively and efficiently. This framework provides a promising direction for improving the stability and accuracy of PINNs-based solvers in scientific computing.
△ Less
Submitted 11 June, 2025; v1 submitted 10 June, 2025;
originally announced June 2025.
-
Nonperturbative Renormalization as Riemann-Hilbert Decomposition of Schwinger-Dyson D-Module
Authors:
Chaoming Song
Abstract:
We present a non-perturbative formulation of renormalization by viewing the regularized Schwinger-Dyson hierarchy as a meromorphic connection, that is, as a D-module on the product of spacetime with the regulator disc. The irregular Riemann-Hilbert correspondence splits this connection into a purely formal submodule that contains every ultraviolet pole and a holomorphic submodule that is finite at…
▽ More
We present a non-perturbative formulation of renormalization by viewing the regularized Schwinger-Dyson hierarchy as a meromorphic connection, that is, as a D-module on the product of spacetime with the regulator disc. The irregular Riemann-Hilbert correspondence splits this connection into a purely formal submodule that contains every ultraviolet pole and a holomorphic submodule that is finite at the regulator origin. In this setting, counterterms coincide with the formal Stokes data, the renormalized theory is identified with the analytic submodule, and the Callan-Symanzik flow appears as an isomonodromic deformation in the physical scale. Viewed through this lens, the Connes-Kreimer construction with its graph Hopf algebra, Bogoliubov recursion, and Birkhoff factorisation is simply the perturbative shadow cast by the global geometric decomposition of the full Schwinger-Dyson system.
△ Less
Submitted 27 April, 2025;
originally announced April 2025.
-
Stochastic quantization of $λφ_2^4$- theory in 2-d Moyal space
Authors:
Chunqiu Song,
Hendrik Weber,
Raimar Wulkenhaar
Abstract:
There is strong evidence for the conjecture that the $λφ^4$ QFT- model on 4-dimensional non-commutative Moyal space can be non-perturbatively constructed. As preparation, in this paper we construct the 2-dimensional case with the method of stochastic quantization. We show the local well-posedness and global well-posedness of the stochastic quantization equation, leading to a construction of the Mo…
▽ More
There is strong evidence for the conjecture that the $λφ^4$ QFT- model on 4-dimensional non-commutative Moyal space can be non-perturbatively constructed. As preparation, in this paper we construct the 2-dimensional case with the method of stochastic quantization. We show the local well-posedness and global well-posedness of the stochastic quantization equation, leading to a construction of the Moyal $λφ^4_2$ measure for any non-negative coupling constant $λ$.
△ Less
Submitted 23 May, 2025; v1 submitted 4 February, 2025;
originally announced February 2025.
-
PINNeik: Eikonal solution using physics-informed neural networks
Authors:
Umair bin Waheed,
Ehsan Haghighat,
Tariq Alkhalifah,
Chao Song,
Qi Hao
Abstract:
The eikonal equation is utilized across a wide spectrum of science and engineering disciplines. In seismology, it regulates seismic wave traveltimes needed for applications like source localization, imaging, and inversion. Several numerical algorithms have been developed over the years to solve the eikonal equation. However, these methods require considerable modifications to incorporate additiona…
▽ More
The eikonal equation is utilized across a wide spectrum of science and engineering disciplines. In seismology, it regulates seismic wave traveltimes needed for applications like source localization, imaging, and inversion. Several numerical algorithms have been developed over the years to solve the eikonal equation. However, these methods require considerable modifications to incorporate additional physics, such as anisotropy, and may even breakdown for certain complex forms of the eikonal equation, requiring approximation methods. Moreover, they suffer from computational bottleneck when repeated computations are needed for perturbations in the velocity model and/or the source location, particularly in large 3D models. Here, we propose an algorithm to solve the eikonal equation based on the emerging paradigm of physics-informed neural networks (PINNs). By minimizing a loss function formed by imposing the eikonal equation, we train a neural network to output traveltimes that are consistent with the underlying partial differential equation. We observe sufficiently high traveltime accuracy for most applications of interest. We also demonstrate how the proposed algorithm harnesses machine learning techniques like transfer learning and surrogate modeling to speed up traveltime computations for updated velocity models and source locations. Furthermore, we use a locally adaptive activation function and adaptive weighting of the terms in the loss function to improve convergence rate and solution accuracy. We also show the flexibility of the method in incorporating medium anisotropy and free-surface topography compared to conventional methods that require significant algorithmic modifications. These properties of the proposed PINN eikonal solver are highly desirable in obtaining a flexible and efficient forward modeling engine for seismological applications.
△ Less
Submitted 17 May, 2021; v1 submitted 16 July, 2020;
originally announced July 2020.
-
Solving the generalized Higgs model from the generalized CRS model
Authors:
Ci Song,
Yan Li,
Jing-Ling Chen
Abstract:
In this paper, we reveal a direct relation between the generalized one-dimensional Carinena-Ranada-Santander (CRS) model and the radial part of two-dimensional generalized Higgs model. By this relation, we construct a series of quasi-exactly solutions for the two-dimensional Higgs model from a solved generalized CRS model.
In this paper, we reveal a direct relation between the generalized one-dimensional Carinena-Ranada-Santander (CRS) model and the radial part of two-dimensional generalized Higgs model. By this relation, we construct a series of quasi-exactly solutions for the two-dimensional Higgs model from a solved generalized CRS model.
△ Less
Submitted 17 December, 2014; v1 submitted 23 September, 2013;
originally announced September 2013.
-
Geometric solitons of Hamiltonian flows on manifolds
Authors:
Chong Song,
Xiaowei Sun,
Youde Wang
Abstract:
It is well-known that the LIE(Locally Induction Equation) admit soliton-type solutions and same soliton solutions arise from different and apparently irrelevant physical models. By comparing the solitons of LIE and Killing magnetic geodesics, we observe that these solitons are essentially decided by two families of isometries of the domain and the target space respectively. With this insight, we p…
▽ More
It is well-known that the LIE(Locally Induction Equation) admit soliton-type solutions and same soliton solutions arise from different and apparently irrelevant physical models. By comparing the solitons of LIE and Killing magnetic geodesics, we observe that these solitons are essentially decided by two families of isometries of the domain and the target space respectively. With this insight, we propose the new concept of geometric solitons of Hamiltonian flows on manifolds, such as geometric Schrödinger flows and KdV flows for maps. Moreover, we give several examples of geometric solitons of the Schrödinger flow and geometric KdV flow, including magnetic curves as geometric Schrödinger solitons and explicit geometric KdV solitons on surfaces of revolution.
△ Less
Submitted 17 December, 2012;
originally announced December 2012.
-
The Quasi-exact models in two-dimensional curved space based on the generalized CRS Harmonic Oscillator
Authors:
Ci Song,
Yan Li,
Jing-Ling Chen
Abstract:
In this paper, by searching the relation between the radial part of Higgs harmonic oscillator in the two-dimensional curved space and the generalized CRS harmonic oscillator model, we can find a series of quasi-exact models in two-dimensional curved space based on this relation.
In this paper, by searching the relation between the radial part of Higgs harmonic oscillator in the two-dimensional curved space and the generalized CRS harmonic oscillator model, we can find a series of quasi-exact models in two-dimensional curved space based on this relation.
△ Less
Submitted 23 March, 2012;
originally announced March 2012.
-
Applications of Polynomial Algebras to Deformed Oscillator
Authors:
Ci Song,
Fu-Lin Zhang,
Jing-Ling Chen
Abstract:
The polynomial algebra is a deformed SU(2) algebra. Here, we use polynomial algebra as a method to solve a series of deformed oscillators. Meanwhile, we find a series of physics systems corresponding with polynomial algebra with different maximal order.
The polynomial algebra is a deformed SU(2) algebra. Here, we use polynomial algebra as a method to solve a series of deformed oscillators. Meanwhile, we find a series of physics systems corresponding with polynomial algebra with different maximal order.
△ Less
Submitted 20 October, 2009; v1 submitted 13 September, 2009;
originally announced September 2009.
