Geometry of linear convolutional networks
From MaRDI portal
Recommendations
- Theory of deep convolutional neural networks. III: Approximating radial functions
- The global optimization geometry of shallow linear neural networks
- Theory of deep convolutional neural networks. II: Spherical analysis
- Optimization Landscape of Neural Networks
- Universality of deep convolutional neural networks
Cites work
- Algebraic statistics
- Algorithms in real algebraic geometry
- Duality of multiple root loci
- Geometric reasoning with logic and algebra
- Learning deep linear neural networks: Riemannian gradient flows and convergence to global minimizers
- Maximum likelihood estimation of the latent class model through model boundary decomposition
- Mixtures and products in two graphical models
- Nonconvex Optimization Meets Low-Rank Matrix Factorization: An Overview
- Optimal Transport to a Variety
- Some remarks on equations defining coincident root loci
- The Euclidean distance degree of an algebraic variety
- The implicit bias of gradient descent on separable data
- Wasserstein distance to independence models
Cited in
(17)- Algebraic complexity and neurovariety of linear convolutional networks
- Processing simple geometric attributes with autoencoders
- Function space and critical points of linear convolutional networks
- Geometrical multi-resolution network based on ridgelet frame
- Geometry of linear neural networks: equivariance and invariance under permutation groups
- Real roots of real cubics and optimization
- Multilinear compressive sensing and an application to convolutional linear networks
- A linear algebra approach to minimal convolutional encoders
- Geometrical aspects of discrimination by multilayer perceptrons
- The geometry of the deep linear network
- Functional dimension of feedforward ReLU neural networks
- Geometry and optimization of shallow polynomial networks
- scientific article; zbMATH DE number 7626727 (Why is no real title available?)
- Side effects of learning from low-dimensional data embedded in a Euclidean space
- Algebraic representations for faster predictions in convolutional neural networks
- How convolutional neural network architecture biases learned opponency and color tuning
- Applied harmonic analysis and data science. Abstracts from the workshop held April 21--26, 2024
This page was built for publication: Geometry of linear convolutional networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5097687)
