Machine Learning Algorithms - Second Edition

Sparse PCA

The scikit-learn library provides different PCA variants that can solve particular problems. I do suggest reading the original documentation. However, I'd like to mention SparsePCA, which allows exploiting the natural sparsity of data while extracting principal components. If you think about the handwritten digits or other images that must be classified, their initial dimensionality can be quite high (a 10 x 10 image has 100 features). However, applying a standard PCA selects only the average most important features, assuming that every sample can be rebuilt using the same components. Simplifying this, this is equivalent to the following:

On the other hand, we can always use a limited number of components, but without the limitation ...