Open AccessOnline Stochastic Principal Component Analysis
Author(s) -
Nuraddeen Adamu,
Samaila Abdullahi,
Sunday Abraham Musa
Publication year - 2022
Publication title -
caliphate journal of science and technology
Language(s) - English
Resource type - Journals
eISSN - 2705-3121
pISSN - 2705-313X
DOI - 10.4314/cajost.v4i1.13
Subject(s) - principal component analysis , subspace topology , block (permutation group theory) , computer science , matlab , pattern recognition (psychology) , dimension (graph theory) , convergence (economics) , artificial intelligence , algorithm , mathematics , geometry , pure mathematics , economics , economic growth , operating system
This paper studied Principal Component Analysis (PCA) in an online. The problem is posed as a subspace optimization problem and solved using gradient based algorithms. One such algorithm is the Variance-Reduced PCA (VR-PCA). The VR-PCA was designed as an improvement to the classical online PCA algorithm known as the Oja’s method where it only handled one sample at a time. The paper developed Block VR-PCA as an improved version of VR-PCA. Unlike prominent VR-PCA, the Block VR-PCA was designed to handle more than one dimension in subspace optimization at a time and it showed good performance. The Block VR-PCA and Block Oja method were compared experimentally in MATLAB using synthetic and real data sets, their convergence results showed Block VR-PCA method appeared to achieve a minimum steady state error than Block Oja method.
Keywords: Online Stochastic; Principal Component Analysis; Block Variance-Reduced; Block Oja