On the Eigenspectrum of the Gram Matrix and the Generalization Error of Kernel-PCA

DOI10.1109/TIT.2005.850052zbMath1310.15076OpenAlexW2139095575MaRDI QIDQ3547627

Christopher K. I. Williams, Nello Cristianini, John Shawe-Taylor, J. S. Kandola

Publication date: 21 December 2008

Published in: IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1109/tit.2005.850052

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Learning and adaptive systems in artificial intelligence (68T05) Eigenvalues, singular values, and eigenvectors (15A18) Random matrices (algebraic aspects) (15B52)

Related Items

Non-asymptotic properties of spectral decomposition of large Gram-type matrices and applications, Neural-network-based approach for extracting eigenvectors and eigenvalues of real normal matrices and some extension to real matrices, Accuracy of suboptimal solutions to kernel principal component analysis, Statistical properties of kernel principal component analysis, Unnamed Item, Concentration of kernel matrices with application to kernel spectral clustering, Two-sample test for equal distributions in separate metric space: New maximum mean discrepancy based approaches, Universally consistent vertex classification for latent positions graphs, Compressive statistical learning with random feature moments, On spectral windows in supervised learning from data, Another neural network based approach for computing eigenvalues and eigenvectors of real skew-symmetric matrices, Some remarks on MCMC estimation of spectra of integral operators, Explicit embeddings for nearest neighbor search with Mercer kernels, Learning noisy linear classifiers via adaptive and selective sampling, Statistical performance of support vector machines, Nonasymptotic upper bounds for the reconstruction error of PCA, Mercer's theorem on general domains: on the interaction between measures, kernels, and RKHSs, Transfer bounds for linear feature learning, Decomposing the tensor kernel support vector machine for neuroscience data with structured labels, The smallest eigenvalues of random kernel matrices: asymptotic results on the min kernel, Robust recovery of multiple subspaces by geometric \(l_{p}\) minimization, On the eigenvector bias of Fourier feature networks: from regression to solving multi-scale PDEs with physics-informed neural networks, Unsupervised slow subspace-learning from stationary processes, Robust dimension-free Gram operator estimates, Statistical Analysis and Parameter Selection for Mapper, Dominated concentration, A spectral graph approach to discovering genetic ancestry, Oracle inequalities for support vector machines that are based on random entropy numbers, Principal component analysis for multivariate extremes, Model reduction and neural networks for parametric PDEs, High-probability bounds for the reconstruction error of PCA, Random discretization of the finite Fourier transform and related kernel random matrices, Statistical analysis of Mapper for stochastic and multivariate filters, Approximate kernel PCA: computational versus statistical trade-off