Cross-view kernel transfer
This addresses kernel matrix completion for multi-view data with missing views, which is incremental as it builds on existing kernel methods for multi-view learning.
The paper tackles the kernel completion problem in multi-view data where samples can be fully missing in some views, proposing Cross-View Kernel Transfer (CVKT) to complete kernel matrices by transforming features from other views using kernel alignment, and demonstrates benefits on simulated, digit, gesture, and biological datasets.
We consider the kernel completion problem with the presence of multiple views in the data. In this context the data samples can be fully missing in some views, creating missing columns and rows to the kernel matrices that are calculated individually for each view. We propose to solve the problem of completing the kernel matrices with Cross-View Kernel Transfer (CVKT) procedure, in which the features of the other views are transformed to represent the view under consideration. The transformations are learned with kernel alignment to the known part of the kernel matrix, allowing for finding generalizable structures in the kernel matrix under completion. Its missing values can then be predicted with the data available in other views. We illustrate the benefits of our approach with simulated data, multivariate digits dataset and multi-view dataset on gesture classification, as well as with real biological datasets from studies of pattern formation in early \textit{Drosophila melanogaster} embryogenesis.