The Geometry Of Kernel Canonical Correlation Analysis

By Malte Kuss. 2003

108 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/40f5d93bf10c285ec45d928d61d4f3c029bfbbfe

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

kernlab - An S4 Package for Kernel Methods in R (2004, methodology): “…CCA can be extended and used in a dot product space F which is related to the input space by a possibly nonlinear map Φ : R N → F , x 7→ X as Following (Kuss and Graepel, 2003), the kernlab implementation of a KCCA projects the data vectors on a new coordinate system using KPCA and uses linear…”
Manifold Relevance Determination (2012, background): “Different approaches exploit different characteristics of the data such as, correlation (Kuss & Graepel, 2003; Ham et al., 2005), or mutual information (Memisevic et al., 2011).”
Factorized Orthogonal Latent Spaces (2010, background): “However, (Archam-Factorized beau and Bach, 2008; Klami and Kaski, 2008) build on probabilistic CCA and therefore do not generalize to non-linear mappings.”
A Kernel Subspace Method by Stochastic Realization for Learning Nonlinear Dynamical Systems (2006, methodology): “Then,Φz with respect to the principal components is given by Cz := ΦzUz = ΦzΦzAz = GzAz [11].”
A Survey of Multi-View Representation Learning (2016, unknown): “Consequently, the statistical properties of KCCA have been investigated from several aspects [2, 93].”
Learning over Sets using Kernel Principal Angles (2003, methodology): “The fact that only input space dimensionRn is used constrains the applicability of the technique to relatively small input sets, and the assumption of a Gaussian distribution limits the kind of variability alongthe input sequence which can be effectively tolerated.”
Factorized Latent Spaces with Structured Sparsity (2010, unknown): “1(a)), such as Canonical Correlation Analysis (CCA) [12, 3], the shared Kernel Information Embedding model (sKIE) [23], and the shared Gaussian Process Latent Variable Model (shared GPLVM) [21, 6, 15], have shown particularly effective to model the dependencies between the modalities.”
Similarity of Neural Network Representations Revisited (2019, methodology): “To improve robustness, singular vector CCA (SVCCA) performs CCA on truncated singular value decompositions of X and Y (Raghu et al., 2017; Mroueh et al., 2015; Kuss & Graepel, 2003).”
Kernel Sliced Inverse Regression with Applications to Classification (2008, background): “Kuss and Graepel (2003) addressed the connection between the kernel Fisher discriminant analysis with the kernel canonical correlation analysis.”