Derivation of Learning Rules for Coupled Principal Component Analysis in a Lagrange-Newton Framework
April 28, 2022 ยท Declared Dead ยท + Add venue
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Ralf Mรถller
arXiv ID
2204.13460
Category
cs.NE: Neural & Evolutionary
Citations
0
Last Checked
4 months ago
Abstract
We describe a Lagrange-Newton framework for the derivation of learning rules with desirable convergence properties and apply it to the case of principal component analysis (PCA). In this framework, a Newton descent is applied to an extended variable vector which also includes Lagrange multipliers introduced with constraints. The Newton descent guarantees equal convergence speed from all directions, but is also required to produce stable fixed points in the system with the extended state vector. The framework produces "coupled" PCA learning rules which simultaneously estimate an eigenvector and the corresponding eigenvalue in cross-coupled differential equations. We demonstrate the feasibility of this approach for two PCA learning rules, one for the estimation of the principal, the other for the estimate of an arbitrary eigenvector-eigenvalue pair (eigenpair).
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
๐ Similar Papers
In the same crypt โ Neural & Evolutionary
๐ฎ
๐ฎ
The Ethereal
R.I.P.
๐ป
Ghosted
Deep Learning using Rectified Linear Units (ReLU)
R.I.P.
๐ป
Ghosted
Generative Adversarial Text to Image Synthesis
R.I.P.
๐ป
Ghosted
Regularized Evolution for Image Classifier Architecture Search
R.I.P.
๐ป
Ghosted
Temporal Ensembling for Semi-Supervised Learning
๐
๐
Old Age
Learning Structured Sparsity in Deep Neural Networks
Died the same way โ ๐ป Ghosted
R.I.P.
๐ป
Ghosted
Federated Learning: Strategies for Improving Communication Efficiency
R.I.P.
๐ป
Ghosted
In-Datacenter Performance Analysis of a Tensor Processing Unit
R.I.P.
๐ป
Ghosted
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning
R.I.P.
๐ป
Ghosted