Sparse and
large-scale learning with heterogeneous data
1) Book chapter: http://cosmal.ucsd.edu/~gert/papers/book.pdf
(it describes applications in bioinformatics and therefore introduces the
machine learning quite smoothly; the details of the biology are not really
relevant, so no need to understand all details there)
2) For the "sparse PCA" part of the talk: http://cosmal.ucsd.edu/~gert/papers/SIAMSparsePCA.pdf
(don't worry if it gets too detailed: it should mainly help to introduce the
problem)