Improving the expressive power of deep neural networks through integral activation transform
From MaRDI portal
(Redirected from Publication:6648522)
Recommendations
- SignReLU neural network and its approximation ability
- Neural networks with ReLU powers need less depth
- Entropy-based guidance of deep neural networks for accelerated convergence and improved performance
- Low dimensional approximation and generalization of multivariate functions on smooth manifolds using deep ReLU neural networks
Cites work
- Bayesian learning for neural networks
- scientific article; zbMATH DE number 1324223 (Why is no real title available?)
- Level Set Learning with Pseudoreversible Neural Networks for Nonlinear Dimension Reduction in Function Approximation
- Nonlinear approximation and (deep) ReLU networks
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Plateau phenomenon in gradient descent training of RELU networks: explanation, quantification, and avoidance
- Transferable neural networks for partial differential equations
Cited in
(2)
This page was built for publication: Improving the expressive power of deep neural networks through integral activation transform
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6648522)
