Canonical Correlation Analysis (CCA) — use case.
hardAnswer
- Finds pairs of linear projections (u, v) of two multivariate datasets X, Y that maximize their correlation.
- Uses: (1) multi-view learning — combine images and text embeddings, (2) genomics (SNPs vs gene expression), (3) recommendation with side information.
- Deep CCA extends with neural nets on each side.
- Foundational for cross-modal alignment.
Check yourself — multiple choice
- Random
- Find linear projections (u, v) of two datasets X, Y that maximize corr(Xu, Yv); multi-view learning, cross-modal alignment; deep CCA extends to NNs
- Same as PCA
- Not real
CCA: max correlation between two views' linear projections; deep CCA extends.
#dimensionality-reduction
Practise Unsupervised Learning
214 interview questions in this topic.