A dive into spectral inference networks: improved algorithms for self-supervised learning of continuous spectral representations

DOI10.1007/S10483-023-2998-7zbMATH Open1515.68272MaRDI QIDQ6132297FDOQ6132297

Authors: J. Wu, Sifan Wang, Paris Perdikaris 

Publication date: 14 July 2023

Published in: AMM. Applied Mathematics and Mechanics. (English Edition) (Search for Journal in Brave)

Recommendations

• Pseudoinverse graph convolutional networks. Fast filters tailored for large eigengaps of dense graphs and hypergraphs
• Spectral learning on matrices and tensors
• Discovering phase field models from image data with the pseudo-spectral physics informed neural networks
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Convergence Rates for Learning Linear Operators from Noisy Data

zbMATH Keywords

neural networkpartial differential equationspectral learningslow features analysis

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Eigenvalue problems for linear operators (47A75) PDEs in connection with computer science (35Q68)

Cites Work

• The Geometry of Algorithms with Orthogonality Constraints
• Numerical methods for large eigenvalue problems
• SPECTRA OF SELF-ADJOINT EXTENSIONS AND APPLICATIONS TO SOLVABLE SCHRÖDINGER OPERATORS
• Title not available (Why is that?)
• Title not available (Why is that?)
• Eigenvalue problems, spectral parameter power series, and modern applications
• Slow Feature Analysis: Unsupervised Learning of Invariances
• The Deep Ritz Method: a deep learning-based numerical algorithm for solving variational problems
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• On the relation of slow feature analysis and Laplacian eigenmaps
• Approximate controllability, exact controllability, and conical eigenvalue intersections for quantum mechanical systems
• Nonlinear eigenproblems in image processing and computer vision
• Solving high-dimensional eigenvalue problems using deep neural networks: a diffusion Monte Carlo like approach
• Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks

Cited In (1)

• Preface: machine-learning approaches for computational mechanics

This page was built for publication: A dive into spectral inference networks: improved algorithms for self-supervised learning of continuous spectral representations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6132297)