Professor Juho Rousu ‘Sparse Non-linear Canonical Correlation Analysis’

Map Unavailable

Date(s) - 02/05/2019
2:00 pm - 3:00 pm

University of Aberdeen

The University of Aberdeen will be hosting Professor Juho Rousu, Professor of Computer Science, Aalto University, Finland on Thursday 2 May and his talk is ‘Sparse Non-linear Canonical Correlation Analysis’.

Canonical correlation analysis (CCA) methods find multivariate relations in two-view data settings. CCA can be seen as a relative of principal component analysis, when the objective is to explain covariance between two views rather than variance within one view. When applying CCA methods, the interest is typically to find the related variables in the two views, and to uncover the relation that couples the variables. In general, it is difficult to achieve these objectives if the underlying multivariate relation is non-linear and the data is high-dimensional. CCA methods in the literature tend to be either non-linear (such as kernel CCA and deep CCA) or sparse (sparse CCA) but not both. In this presentation, I discuss recent progress in learning CCA models where the underlying relations are both non-linear and sparse. In general, the methods are based on mapping a sparse projection of the data through a non-linear kernel function. This approach can be used to uncover both within-view non-linearities (similar to kernel CCA and deep CCA) but also between-view non-linearities.

Andrew, G., Arora, R., Bilmes, J. and Livescu, K., 2013, February. Deep canonical correlation analysis. In International conference on machine learning (pp. 1247-1255).
Hardoon, D.R., Szedmak, S. and Shawe-Taylor, J., 2004. Canonical correlation analysis: An overview with application to learning methods. Neural computation, 16(12), pp.2639-2664.
Uurtio, V., Bhadra, S. and Rousu, J., 2018, November. Sparse Non-linear CCA through Hilbert-Schmidt Independence Criterion. In 2018 IEEE International Conference on Data Mining (ICDM) (pp. 1278-1283). IEEE.
Uurtio, V., Bhadra, S. and Rousu, J., 2019, Large-scale sparse kernel canonical correlation analysis, in review
Uurtio, V., Monteiro, J.M., Kandola, J., Shawe-Taylor, J., Fernandez-Reyes, D. and Rousu, J., 2018. A tutorial on canonical correlation methods. ACM Computing Surveys (CSUR), 50(6), p.95

Short Bio:
Juho Rousu is a Professor of Computer Science at Aalto University, Finland. Rousu obtained his PhD in 2001 form University of Helsinki, while working at VTT Technical Centre of Finland. In 2003-2005 he was a Marie Curie Fellow at Royal Holloway University of London. In 2005-2011 he held Lecturer and Professor positions at University of Helsinki, before moving to Aalto University in 2012 where he leads a research group on Kernel Methods, Pattern Analysis and Computational Metabolomics (KEPACO). Rousu’s main research interest is in learning with multiple and structured targets, multiple views and ensembles, with methodological emphasis in regularised learning, kernels and sparsity, as well as efficient convex/non-convex optimisation methods. His applications of interest include metabolomics, biomedicine, pharmacology and synthetic biology.

Professor Rousu is hosted by Dr Wei Pang whilst he is visiting Aberdeen.

This entry was posted in .