On the Exact Computation of Linear Frequency Principle Dynamics and Its Generalization

DOI10.1137/21M1444400OpenAlexW3092919487MaRDI QIDQ5051354

Zhi-Qin John Xu, Yaoyu Zhang, Zheng Ma, Tao Luo

Publication date: 23 November 2022

Published in: SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2010.08153

zbMATH Keywords

optimizationFourier analysistwo-layer neural networkdeep learningneural tangent kernelfrequency principle

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Function spaces arising in harmonic analysis (42B35)

Cites Work

• Unnamed Item
• Machine learning from a continuous viewpoint. I
• Smoothing noisy data with spline functions: Estimating the correct degree of smoothing by the method of generalized cross-validation
• A comparative analysis of optimization and generalization properties of two-layer neural network and random feature models under gradient descent dynamics
• Adaptive activation functions accelerate convergence in deep and physics-informed neural networks
• The vectorization of ITPACK 2C
• A Phase Shift Deep Neural Network for High Frequency Approximation and Wave Problems
• Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
• Multi-Scale Deep Neural Network (MscaleDNN) for Solving Poisson-Boltzmann Equation in Complex Domains
• Approximation by superpositions of a sigmoidal function