Autoencoder Neural Network Architecture
An autoencoder is a neural network that’s composed of two main components, the encoder and the decoder layers. Autoencoders works by compressing the original data into a lower-dimensional vector representation...
Matrix Operations and Neural Networks
A video by Luis Serrano provides an introduction to recurrent neural networks, including the mathematical representations of neural networks using linear algebra. This article also provides some example of using...
Restricted Boltzmann Machine
Boltzmann machine is a type of neural network which is inspired by the work of Ludwig Boltzmann in the field of statistical mechanics. We’re specifically looking at a version of...
Matrix Eigendecomposition
Eigendecomposition is the factorization of a matrix, where the matrix is represented by its eigenvalues and eigenvectors. Suppose we have the matrix \(A\) whose dimension is \(n \times n\), \(A\)...
Singular Value Decomposition
As described by Professor Gilbert Strang from the Massachusetts Institute of Technology in this video, every \(m \times n\) matrix \(A\) can be factored as follows. \[A = U \Sigma...