Variable Star Classification Using Multi-View Metric Learning
This work addresses star classification for astronomy, but it is incremental as it extends standard multi-view learning to matrix-variate cases.
The paper tackles the problem of classifying variable stars by introducing a multi-view metric learning framework that learns to discriminate in a multi-faceted feature space, eliminating the need to pre-combine features, and demonstrates favorable performance on UCR Starlight and LINEAR datasets.
Our multi-view metric learning framework enables robust characterization of star categories by directly learning to discriminate in a multi-faceted feature space, thus, eliminating the need to combine feature representations prior to fitting the machine learning model. We also demonstrate how to extend standard multi-view learning, which employs multiple vectorized views, to the matrix-variate case which allows very novel variable star signature representations. The performance of our proposed methods is evaluated on the UCR Starlight and LINEAR datasets. Both the vector and matrix-variate versions of our multi-view learning framework perform favorably --- demonstrating the ability to discriminate variable star categories.