Quantifying Training Difficulty and Accelerating Convergence in Neural Network-Based PDE Solvers

• URL: http://arxiv.org/abs/2410.06308v1
• Date: Tue, 8 Oct 2024 19:35:19 GMT
• Title: Quantifying Training Difficulty and Accelerating Convergence in Neural Network-Based PDE Solvers
• Authors: Chuqi Chen, Qixuan Zhou, Yahong Yang, Yang Xiang, Tao Luo,
• Abstract summary: We investigate the training dynamics of neural network-based PDE solvers. We find that two techniques, partition of unity (PoU) and variance scaling (VS) enhance the effective rank. Experiments using popular PDE-solving frameworks, such as PINN, Deep Ritz, and the operator learning framework DeepOnet, confirm that these techniques consistently speed up convergence.
• Score: 9.936559796069844
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Neural network-based methods have emerged as powerful tools for solving partial differential equations (PDEs) in scientific and engineering applications, particularly when handling complex domains or incorporating empirical data. These methods leverage neural networks as basis functions to approximate PDE solutions. However, training such networks can be challenging, often resulting in limited accuracy. In this paper, we investigate the training dynamics of neural network-based PDE solvers with a focus on the impact of initialization techniques. We assess training difficulty by analyzing the eigenvalue distribution of the kernel and apply the concept of effective rank to quantify this difficulty, where a larger effective rank correlates with faster convergence of the training error. Building upon this, we discover through theoretical analysis and numerical experiments that two initialization techniques, partition of unity (PoU) and variance scaling (VS), enhance the effective rank, thereby accelerating the convergence of training error. Furthermore, comprehensive experiments using popular PDE-solving frameworks, such as PINN, Deep Ritz, and the operator learning framework DeepOnet, confirm that these initialization techniques consistently speed up convergence, in line with our theoretical findings.

Related papers

• Improving PINNs By Algebraic Inclusion of Boundary and Initial Conditions [0.1874930567916036]
  "AI for Science" aims to solve fundamental scientific problems using AI techniques. In this work we explore the possibility of changing the model being trained from being just a neural network to being a non-linear transformation of it. This reduces the number of terms in the loss function than the standard PINN losses.
  arXiv  Detail & Related papers  (2024-07-30T11:19:48Z)
• VS-PINN: A fast and efficient training of physics-informed neural networks using variable-scaling methods for solving PDEs with stiff behavior [0.0]
  We propose a new method for training PINNs using variable-scaling techniques. We will demonstrate the effectiveness of the proposed method for these problems and confirm that it can significantly improve the training efficiency and performance of PINNs.
  arXiv  Detail & Related papers  (2024-06-10T14:11:15Z)
• Efficient Neural PDE-Solvers using Quantization Aware Training [71.0934372968972]
  We show that quantization can successfully lower the computational cost of inference while maintaining performance. Our results on four standard PDE datasets and three network architectures show that quantization-aware training works across settings and three orders of FLOPs magnitudes.
  arXiv  Detail & Related papers  (2023-08-14T09:21:19Z)
• Implicit Stochastic Gradient Descent for Training Physics-informed Neural Networks [51.92362217307946]
  Physics-informed neural networks (PINNs) have effectively been demonstrated in solving forward and inverse differential equation problems. PINNs are trapped in training failures when the target functions to be approximated exhibit high-frequency or multi-scale features. In this paper, we propose to employ implicit gradient descent (ISGD) method to train PINNs for improving the stability of training process.
  arXiv  Detail & Related papers  (2023-03-03T08:17:47Z)
• Ensemble learning for Physics Informed Neural Networks: a Gradient Boosting approach [10.250994619846416]
  We present a new training paradigm referred to as "gradient boosting" (GB) Instead of learning the solution of a given PDE using a single neural network directly, our algorithm employs a sequence of neural networks to achieve a superior outcome. This work also unlocks the door to employing ensemble learning techniques in PINNs.
  arXiv  Detail & Related papers  (2023-02-25T19:11:44Z)
• Tunable Complexity Benchmarks for Evaluating Physics-Informed Neural Networks on Coupled Ordinary Differential Equations [64.78260098263489]
  In this work, we assess the ability of physics-informed neural networks (PINNs) to solve increasingly-complex coupled ordinary differential equations (ODEs) We show that PINNs eventually fail to produce correct solutions to these benchmarks as their complexity increases. We identify several reasons why this may be the case, including insufficient network capacity, poor conditioning of the ODEs, and high local curvature, as measured by the Laplacian of the PINN loss.
  arXiv  Detail & Related papers  (2022-10-14T15:01:32Z)
• Neural Basis Functions for Accelerating Solutions to High Mach Euler Equations [63.8376359764052]
  We propose an approach to solving partial differential equations (PDEs) using a set of neural networks. We regress a set of neural networks onto a reduced order Proper Orthogonal Decomposition (POD) basis. These networks are then used in combination with a branch network that ingests the parameters of the prescribed PDE to compute a reduced order approximation to the PDE.
  arXiv  Detail & Related papers  (2022-08-02T18:27:13Z)
• Multi-resolution partial differential equations preserved learning framework for spatiotemporal dynamics [11.981731023317945]
  Physics-informed deep learning (PiDL) addresses these challenges by incorporating physical principles into the model. We propose to leverage physics prior knowledge by baking'' the discretized governing equations into the neural network architecture. This method, embedding discretized PDEs through convolutional residual networks in a multi-resolution setting, largely improves the generalizability and long-term prediction.
  arXiv  Detail & Related papers  (2022-05-09T01:27:58Z)
• Learning Physics-Informed Neural Networks without Stacked Back-propagation [82.26566759276105]
  We develop a novel approach that can significantly accelerate the training of Physics-Informed Neural Networks. In particular, we parameterize the PDE solution by the Gaussian smoothed model and show that, derived from Stein's Identity, the second-order derivatives can be efficiently calculated without back-propagation. Experimental results show that our proposed method can achieve competitive error compared to standard PINN training but is two orders of magnitude faster.
  arXiv  Detail & Related papers  (2022-02-18T18:07:54Z)
• On Theory-training Neural Networks to Infer the Solution of Highly Coupled Differential Equations [0.0]
  We present insights into theory-training networks for learning the solution of highly coupled differential equations. We introduce a theory-training technique that, by leveraging regularization, eliminates those oscillations, decreases the final training loss, and improves the accuracy of the inferred solution.
  arXiv  Detail & Related papers  (2021-02-09T15:45:08Z)
• Large-scale Neural Solvers for Partial Differential Equations [48.7576911714538]
  Solving partial differential equations (PDE) is an indispensable part of many branches of science as many processes can be modelled in terms of PDEs. Recent numerical solvers require manual discretization of the underlying equation as well as sophisticated, tailored code for distributed computing. We examine the applicability of continuous, mesh-free neural solvers for partial differential equations, physics-informed neural networks (PINNs) We discuss the accuracy of GatedPINN with respect to analytical solutions -- as well as state-of-the-art numerical solvers, such as spectral solvers.
  arXiv  Detail & Related papers  (2020-09-08T13:26:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.