Multiview Representation Learning for a Union of Subspaces
This work addresses multiview representation learning for data with multiple subspaces, but it is incremental as it builds on existing CCA frameworks with simple heuristics.
The paper tackles the problem of learning multiview mixture models by extending canonical correlation analysis (CCA) to handle a union of subspaces, resulting in improvements over standard CCA on downstream tasks as measured by experimental results.
Canonical correlation analysis (CCA) is a popular technique for learning representations that are maximally correlated across multiple views in data. In this paper, we extend the CCA based framework for learning a multiview mixture model. We show that the proposed model and a set of simple heuristics yield improvements over standard CCA, as measured in terms of performance on downstream tasks. Our experimental results show that our correlation-based objective meaningfully generalizes the CCA objective to a mixture of CCA models.