Constrained Linear Data-feature Mapping for Image Classification
arXiv:1911.10428
Abstract
In this paper, we propose a constrained linear data-feature mapping model as an interpretable mathematical model for image classification using convolutional neural network (CNN) such as the ResNet. From this viewpoint, we establish the detailed connections in a technical level between the traditional iterative schemes for constrained linear system and the architecture for the basic blocks of ResNet. Under these connections, we propose some natural modifications of ResNet type models which will have less parameters but still maintain almost the same accuracy as these corresponding original models. Some numerical experiments are shown to demonstrate the validity of this constrained learning data-feature mapping assumption.
15 page, 2 figures
References in corpus (13)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- FractalNet: Ultra-Deep Neural Networks without Residuals
- Visualizing the Loss Landscape of Neural Nets
- The Reversible Residual Network: Backpropagation Without Storing Activations
- MgNet: A Unified Framework of Multigrid and Convolutional Neural Network
- Multi-level Residual Networks from Dynamical Systems View
- Learning Neural PDE Solvers with Convergence Guarantees
- Learning across scales - A multiscale method for Convolution Neural Networks
- SCAN: A Scalable Neural Networks Framework Towards Compact and Efficient Models
- Deep Multigrid: learning prolongation and restriction matrices
- Optimization Algorithm Inspired Deep Neural Network Structure Design
- A Flow Model of Neural Networks
- Multigrid Neural Architectures