paper-with-me

Papers

Two-level overlapping additive Schwarz preconditioner for training scientific machine learning applications

2024-06-16 · Youngkyu Lee, Alena Kopaničáková, George Em Karniadakis

We introduce a novel two-level overlapping additive Schwarz preconditioner for accelerating the training of scientific machine learning applications. The design of the proposed preconditioner is motivated by the nonlinear two-level overlapping additive Schwarz preconditioner. The neural network parameters are decomposed into groups (subdomains) with overlapping regions. In addition, the network's feed-forward structure is indirectly imposed through a novel subdomain-wise synchronization strategy and a coarse-level training step. Through a series of numerical experiments, which consider physics-informed neural networks and operator learning approaches, we demonstrate that the proposed two-level preconditioner significantly speeds up the convergence of the standard (LBFGS) optimizer while also yielding more accurate machine learning models. Moreover, the devised preconditioner is designed to take advantage of model-parallel computations, which can further reduce the training time.

📄 PDF Abstract BibTeX arXiv:2406.10997

Code (0)

등록된 구현이 없습니다.

Tasks

Operator learning

Similar Papers 제목 키워드 기반

Enhancing training of physics-informed neural networks using domain-decomposition based preconditioning strategies

2023-06-30 · Alena Kopaničáková, Hardik Kothari, George Em Karniadakis, Rolf Krause

We propose to enhance the training of physics-informed neural networks (PINNs). To this aim, we introduce nonlinear additive and multiplicative preconditioning strategies for the widely used L-BFGS optimizer. The nonline…

Gaussian Processes Sampling with Sparse Grids under Additive Schwarz Preconditioner

2024-08-01

Gaussian processes (GPs) are widely used in non-parametric Bayesian modeling, and play an important role in various statistical and machine learning applications. In a variety tasks of uncertainty quantification, generat…

A Non-Monotone Preconditioned Trust-Region Method for Neural Network Training

2026-05-14 · Andrea Angino, Bindi Çapriqi, Shega Likaj, Ken Trotti 외 arxiv

Training deep neural networks at scale can benefit from domain decomposition, where the network is split into subdomains trained in parallel and coupled by a global trust-region mechanism. Building on the Additively Prec…

An Efficient Multilevel Preconditioned Nonlinear Conjugate Gradient Method for Incremental Potential Contact

2026-04-21 · Yu Zhang, Xing Shen, Kemeng Huang, Wei Chen 외 arxiv

Incremental Potential Contact (IPC) guarantees intersection-free simulation but suffers from high computational costs due to the expensive Hessian assembly and linear solves required by Newton's method. While Preconditio…

Hierarchical Attention via Domain Decomposition

2026-06-16 · Stephan Köhler, Oliver Rheinbach arxiv

We propose a hierarchical attention mechanism based on two-level overlapping Schwarz domain decomposition. The method is motivated by the observation that two-level Schwarz domain decomposition methods combine local subd…