Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This lecture covers the solutions to exercises on Canonical Correlation Analysis (CCA), focusing on finding pairs of projection vectors with maximum correlation. It includes manual determination of directions found by CCA, contrasting with PCA, and analyzing Gram matrices for collinear eigenvectors. The lecture also explores the effect of changing kernel width in CCA and discusses the distribution of points in kernel matrices generated by polynomial kernels. Additionally, it delves into the properties of vectors generated by homogeneous polynomial kernels, emphasizing the orthogonality of vectors in 2D space.
This video is available exclusively on Mediaspace for a restricted audience. Please log in to MediaSpace to access it if you have the necessary permissions.
Watch on Mediaspace