PottsMGNet: a mathematical explanation of encoder-decoder based neural networks
From MaRDI portal
Probability in computer science (algorithm analysis, random structures, phase transitions, etc.) (68Q87) Artificial neural networks and deep learning (68T07) Computing methodologies for image processing (68U10) Lattice systems (Ising, dimer, Potts, etc.) and systems on graphs arising in equilibrium statistical mechanics (82B20) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)
Recommendations
- Connections between Operator-Splitting Methods and Deep Neural Networks with Applications in Image Segmentation
- Deep neural networks motivated by partial differential equations
- Rate-optimal denoising with deep neural networks
- MgNet: a unified framework of multigrid and convolutional neural network
- The Modern Mathematics of Deep Learning
Cites work
- A Color Elastica Model for Vector-Valued Image Regularization
- A fast operator-splitting method for Beltrami color image denoising
- A finite element/operator-splitting method for the numerical solution of the three dimensional Monge-Ampère equation
- A finite element/operator-splitting method for the numerical solution of the two dimensional elliptic Monge-Ampère equation
- A New Operator Splitting Method for the Euler Elastica Model for Image Smoothing
- A non-local anisotropic model for phase transitions: asymptotic behaviour of rescaled energies
- A parallel splitting-up method for partial differential equations and its applications to Navier-Stokes equations
- A Penalization-Regularization-Operator Splitting Method for Eikonal Based Traveltime Tomography
- A proposal on machine learning via dynamical systems
- Active contours without edges
- An Edge-Weighted Centroidal Voronoi Tessellation Model for Image Segmentation
- An efficient iterative method for reconstructing surface from point clouds
- An Operator-Splitting Method for the Gaussian Curvature Regularization Model with Applications to Surface Smoothing and Imaging
- Curvature Regularized Surface Reconstruction from Point Clouds
- Deep convolutional neural networks with spatial regularization, volume and star-shape priors for image segmentation
- Deep learning as optimal control problems: models and numerical methods
- Deep neural networks motivated by partial differential equations
- Efficient and convergent preconditioned ADMM for the Potts models
- Fast operator splitting methods for obstacle problems
- Fast operator-splitting algorithms for variational imaging models: some recent developments
- General relaxation methods for initial-value problems with application to multistep schemes
- Global and uniform convergence of subspace correction methods for some convex optimization problems
- Global minimization for continuous multiphase partitioning problems using a dual approach
- scientific article; zbMATH DE number 3833218 (Why is no real title available?)
- scientific article; zbMATH DE number 996056 (Why is no real title available?)
- scientific article; zbMATH DE number 45081 (Why is no real title available?)
- scientific article; zbMATH DE number 671790 (Why is no real title available?)
- scientific article; zbMATH DE number 1982607 (Why is no real title available?)
- scientific article; zbMATH DE number 3076589 (Why is no real title available?)
- Iterative Methods by Space Decomposition and Subspace Correction
- Machine learning from a continuous viewpoint. I
- MgNet: a unified framework of multigrid and convolutional neural network
- Multilayer feedforward networks are universal approximators
- Neural ODE Control for Classification, Approximation, and Transport
- New region force for variational models in image segmentation and high dimensional data clustering
- On the Douglas-Rachford splitting method and the proximal point algorithm for maximal monotone operators
- Opening the black box: low-dimensional dynamics in high-dimensional recurrent neural networks
- Operator splitting algorithms for free surface flows: application to extrusion processes
- Rate of convergence for some constraint decomposition methods for nonlinear variational inequalities
- Short-time heat flow and functions of bounded variation in \(\mathbb R^N\)
- Some Facts About Operator-Splitting and Alternating Direction Methods
- Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
- Structure-preserving deep learning
- Threshold dynamics for networks with arbitrary surface tensions
- Universality of deep convolutional neural networks
Cited in
(19)- Connections between Operator-Splitting Methods and Deep Neural Networks with Applications in Image Segmentation
- Inverse evolution layers: physics-informed regularizers for image segmentation
- Double-well net for image segmentation
- Two-stage selective segmentation method based on exponential weighted geodesic distance driven model and thresholding method
- ReLU-like non-monotonic smooth activation functions based on regularized Heaviside functions and extensions
- A mathematical explanation of UNet
- A trainable variational Chan-Vese network based on algorithm unfolding for image segmentation
- Euler's elastica-based cartoon-smooth-texture image decomposition
- Regularized CNN with geodesic active contour and edge predictor for image segmentation
- An interpretable neural network for diffeomorphic image registration: theory and application
- Diffusion-shock PDEs for deep learning on position-orientation space
- Topology-guaranteed image segmentation: enforcing connectivity, genus, and width constraints
- A mathematical explanation of transformers
- Entropic nonlocal TV with learnable self-similarity in multi-scale deep feature space for image denoising
- A deep layer limit analysis of transformer
- PottsSAGNet: a Potts model-based segmentation network with self-attention and Gaussian regularization
- A unified variational framework for deep weakly supervised image segmentation
- Deep-layer limit and stability analysis of the basic forward-backward-splitting induced network. I: Feed-forward systems
- Robust variational model based tailored UNet: leveraging edge detector and mean curvature for improved image segmentation
This page was built for publication: PottsMGNet: a mathematical explanation of encoder-decoder based neural networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6541918)
