Computer Science – Learning
Scientific paper
2010-08-25
NIPS, Canada (2010)
Computer Science
Learning
Scientific paper
Sparse methods for supervised learning aim at finding good linear predictors from as few variables as possible, i.e., with small cardinality of their supports. This combinatorial selection problem is often turned into a convex optimization problem by replacing the cardinality function by its convex envelope (tightest convex lower bound), in this case the L1-norm. In this paper, we investigate more general set-functions than the cardinality, that may incorporate prior knowledge or structural constraints which are common in many applications: namely, we show that for nondecreasing submodular set-functions, the corresponding convex envelope can be obtained from its \lova extension, a common tool in submodular analysis. This defines a family of polyhedral norms, for which we provide generic algorithmic tools (subgradients and proximal operators) and theoretical results (conditions for support recovery or high-dimensional inference). By selecting specific submodular functions, we can give a new interpretation to known norms, such as those based on rank-statistics or grouped norms with potentially overlapping groups; we also define new norms, in particular ones that can be used as non-factorial priors for supervised learning.
No associations
LandOfFree
Structured sparsity-inducing norms through submodular functions does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Structured sparsity-inducing norms through submodular functions, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Structured sparsity-inducing norms through submodular functions will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-525072