Local Canonical Correlation Analysis for Nonlinear Common Variables Discovery
This work addresses the challenge of multimodal data integration for researchers in machine learning and data analysis, though it appears incremental as an extension of existing CCA and manifold learning techniques.
The paper tackles the problem of discovering hidden common variables from multimodal, nonlinear high-dimensional data by proposing a metric based on local canonical correlation analysis integrated into kernel-based manifold learning, and experimental results confirm that the method successfully identifies these common variables without requiring rigid prior model assumptions.
In this paper, we address the problem of hidden common variables discovery from multimodal data sets of nonlinear high-dimensional observations. We present a metric based on local applications of canonical correlation analysis (CCA) and incorporate it in a kernel-based manifold learning technique.We show that this metric discovers the hidden common variables underlying the multimodal observations by estimating the Euclidean distance between them. Our approach can be viewed both as an extension of CCA to a nonlinear setting as well as an extension of manifold learning to multiple data sets. Experimental results show that our method indeed discovers the common variables underlying high-dimensional nonlinear observations without assuming prior rigid model assumptions.