Jethro's Braindump

Canonical Correlation Analysis

Canonical Correlation Analysis computes a linear projection which maximizes the correlation between two random variables andn enforces orthogonality in the new space. This is commonly used in multi-modal representation learning, where the random variables are modalities.

Extensions of CCA include Kernel CCA, which uses a non-linear projection, and Deep CCA, which addresses the scalability issues of KCCA.

Icon by Laymik from The Noun Project. Website built with ♥ with Org-mode, Hugo, and Netlify.