Singular value decomposition (SVD) is a popular method of matrix decomposition that reduces a complex matrix into its constituent parts so that subsequent matrix calculations are simpler.
The general equation for SVD for real-valued matrices is:
A = U * Σ * V^T
Where A represents the real m x n matrix that is being decomposed, U is a m x m matrix, Σ is an m x n diagonal matrix, and V^T (i.e. V transposed) is the transpose of an n x n matrix.
The columns of the U matrix are called the left-singular vectors of matrix A, while the columns of the V matrix are called the right-singular vectors of A. The diagonal values of the Σ matrix are the singular values of matrix A.
Singular Value Decomposition in Machine Learning
SVD is a commonly used method for data reduction in machine learning, particularly in unsupervised learning algorithms. It's one of the core elements of the recommender systems of international companies like Google, Facebook, Netflix and Youtube, helping them determine the order that pages appear in search results, what content and ads you see, and what shows / movies / videos you might enjoy.
A Gentle Introduction to Singular-Value Decomposition for Machine Learning
Foundations of Machine Learning : Singular Value Decomposition (SVD)
Singular Value Decomposition (SVD) tutorial
Documentaries, videos and podcasts
Singular Value Decomposition (SVD) - MIT OpenCourseWare
May 6th, 2016
- Unsupervised learningA branch of machine learning that tries to make sense of data that has not been labeled, classified, or categorized by extracting features and patterns on its own.
- Non-negative matrix factorization (NMF)A matrix factorization method where all of the values in matrices are constrained to be non-negative so that they are easier to inspect. It is useful in data mining because it has the effect of clustering the input data.
- Nonlinear dimensionality reduction (NDR or NLDR)A process of mapping higher-dimensional data into a lower-dimensional non-linear manifold within higher-dimensional space so that the data can be more easily visualized and interpreted.
- Principal component analysis (PCA)Technique used to find the most valuable parts of all of the variables in a dataset and to then transform the original variables into a smaller set of linear combinations.
- PageRankAn algorithm for calculating the authority of a web page based on link structure.