no code implementations • 1 Apr 2024 • Yue Sun, Chao Chen, Yuesheng Xu, Sihong Xie, Rick S. Blum, Parv Venkitasubramaniam
We theoretically derive conditions where GCNs incorporating such domain differential equations are robust to mismatched training and testing data compared to baseline domain agnostic models.
no code implementations • 5 Mar 2024 • Rui Wang, Yuesheng Xu, Mingsong Yan
The representer theorems unfold that solutions of these learning models can be expressed as linear combination of a finite number of kernel sessions determined by given data and the reproducing kernel.
no code implementations • 13 Jan 2024 • Jie Jiang, Yuesheng Xu
We first developed a numerical method for solving the equation with DNNs as an approximate solution by designing a numerical quadrature that tailors to computing oscillatory integrals involving DNNs.
no code implementations • 14 Sep 2023 • Yuesheng Xu, Taishan Zeng
We develop in this paper a multi-grade deep learning method for solving nonlinear partial differential equations (PDEs).
no code implementations • 2 Jun 2023 • Yuesheng Xu, Haizhang Zhang
We consider deep neural networks with a Lipschitz continuous activation function and with weight matrices of variable widths.
no code implementations • 21 May 2023 • Rui Wang, Yuesheng Xu, Mingsong Yan
Sparsity of a learning solution is a desirable feature in machine learning.
no code implementations • 13 May 2023 • Yuesheng Xu
The MGDL model learns a DNN in several grades, in each of which one constructs a shallow DNN consisting of a relatively small number of layers.
no code implementations • 1 Feb 2023 • Yuesheng Xu
Inspired by the human education process which arranges learning in grades, we propose a multi-grade learning model: We successively solve a number of optimization problems of small sizes, which are organized in grades, to learn a shallow neural network for each grade.
no code implementations • 27 Jul 2022 • Yuesheng Xu, Taishan Zeng
Noting that DNNs have an intrinsic multi-scale structure which is favorable for adaptive representation of functions, by employing a penalty with multiple parameters, we develop DNNs with a multi-scale sparse regularization (SDNN) for effectively representing functions having certain singularities.
no code implementations • 13 May 2022 • Wentao Huang, Yuesheng Xu, Haizhang Zhang
In this current work, we study the convergence of deep neural networks as the depth tends to infinity for two other important activation functions: the leaky ReLU and the sigmoid function.
no code implementations • 28 Sep 2021 • Yuesheng Xu, Haizhang Zhang
Based on the conditions, we present sufficient conditions for piecewise convergence of general deep ReLU networks with increasing widths, and as well as pointwise convergence of deep ReLU convolutional neural networks.
no code implementations • 27 Jul 2021 • Yuesheng Xu, Haizhang Zhang
We explore convergence of deep neural networks with the popular ReLU activation function, as the depth of the networks tends to infinity.
no code implementations • 13 Jun 2019 • Ida Häggström, Yizun Lin, Si Li, Andrzej Krol, Yuesheng Xu, C. Ross Schmidtlein
For the cardiac/lung phantom, an additional cardiac gated 2D-OSEM set was reconstructed.