The SParse Modeling Software developped by Julien Mairal, with contributions of Rodolphe Jenatton is an optimization toolbox providing efficient implementations of algorithms to solve learning problems regularized by the l1 norms and other structured sparsity inducing norms, as well as dictionary learning and matrix factorization. It provides in particular implementations for a significant part of the work on sparse methods conducted in Sierra.


SecureGenome is a software developed by Sriram Sankararaman as a result of our joint work on Privacy in Genomics. In the context of a genomewide association study, it may be possible to detect if a particular individual was part of one of the pools by comparing his or her genome with the summary allele frequencies of the study. The SecureGenome program can be used to choose a set of exposed SNPs so that the ability to detect the presence of individual genotypes in the pooled summary data - as measured by the power of any possible statistical test - remains limited.

SMO for multiple kernel learning

Matlab implementation of the SMO algorithm proposed by Francis Bach and Gert Lanckriet to solve the Support Kernel Machine (SKM) optimization aka Multiple Kernel Learning. See:
Bach, F.R. Lanckriet, G.R.G., Jordan, M.I. (2004). Fast Kernel Learning using Sequential Minimal Optimization . Technical Report CSD-04-1307, Division of Computer Science, University of California, Berkeley

Feature selection in multi-task learning

Code available upon request