We examine Deep Canonically Correlated LSTMs as a way to learn nonlinear transformations of variable length sequences and embed them into a correlated, fixed dimensional space. We use LSTMs to transform multi-view time-series data non-linearly while learning temporal relationships within the data. We then perform correlation analysis on the outputs of these neural networks to find a correlated subspace through which we get our final representation via projection. This work follows from previous work done on Deep Canonical Correlation (DCCA), in which deep feed-forward neural networks were used to learn nonlinear transformations of data while maximizing correlation.
Cite
@article{arxiv.1801.05407,
title = {Deep Canonically Correlated LSTMs},
author = {Neil Mallinar and Corbin Rosset},
journal= {arXiv preprint arXiv:1801.05407},
year = {2018}
}
Comments
8 pages, 3 figures, accepted as the undergraduate honors thesis for Neil Mallinar by The Johns Hopkins University