Original Contribution: Principal components, minor components, and linear neural networks

By Erkki Oja. 1992

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

View-tolerant face recognition and Hebbian learning imply mirror-symmetric neural tuning to head orientation (2016, positive): “In the following we consider as an example a simple Hebbian learning scheme called Oja’s rule [37, 38].”
Incremental Slow Feature Analysis: Adaptive Low-Complexity Slow Feature Updating from High-Dimensional Input Streams (2012, methodology): “Minor components analysis (MCA) (Oja, 1992) incrementally extracts principal components but with a reversed preference: it finds it easiest to extract the components with the smallest eigenvalues.”
Neural Network Approaches to Image Compression (1995, unknown): “This rule has been shown to converge to the largest principal component of the input x [28], [29], [30], [31].”
Adaptive blind signal and image processing (2002, methodology): “The stochastic Gradient Ascent (SGA) proposed by Oja [910] can be formulated as”
Fast and robust fixed-point algorithms for independent component analysis (1999, methodology): “Several principles and methods have been developed to find such a linear representation, including principal component analysis [30], factor analysis [15], projection pursuit [12], [16], independent component analysis [27], etc.”
Iterative kernel principal component analysis for image modeling (2005, background): “Later authors [ 37 , 51] noticed that his convergence proof”
A Unifying Review of Deep and Shallow Anomaly Detection (2020, methodology): “If linear networks are used, such an AE, in fact, recovers the same optimal subspace as spanned by the PCA eigenvectors [382], [383].”
Simple, Efficient, and Neural Algorithms for Sparse Coding (2015, background): “There are also generalizations to computing other principle components Oja (1992).”
Generalizations of principal component analysis, optimization problems, and neural networks (1995, methodology): “• Oja's PCA subspace rule [i.e., eqn (20)] . • Robust variance maximization algorithm (9) with the nonlinearity g(t) = tanh(t) . • Optimal robust error minimization algorithm (17) with the nonlinearity g(t) = log(1 + 5t). [This worked somewhat better than the tanh(t) nonlinearity.] • Approximative r”