You are here: MIMS > EPrints
MIMS EPrints

2013.14: Living on the edge: A geometric theory of phase transitions in convex optimization

2013.14: Dennis Amelunxen, Martin Lotz, Michael B. Mccoy and Joel A. Tropp (2013) Living on the edge: A geometric theory of phase transitions in convex optimization.

Full text available as:

PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
1502 Kb

Abstract

Recent empirical research indicates that many convex optimization problems with random constraints exhibit a phase transition as the number of constraints increases. For example, this phenomenon emerges in the $\ell_1$ minimization method for identifying a sparse vector from random linear samples. Indeed, this approach succeeds with high probability when the number of samples exceeds a threshold that depends on the sparsity level; otherwise, it fails with high probability.

This paper provides the first rigorous analysis that explains why phase transitions are ubiquitous in random convex optimization problems. It also describes tools for making reliable predictions about the quantitative aspects of the transition, including the location and the width of the transition region. These techniques apply to regularized linear inverse problems with random measurements, to demixing problems under a random incoherence model, and also to cone programs with random affine constraints.

These applications depend on foundational research in conic geometry. This paper introduces a new summary parameter, called the statistical dimension, that canonically extends the dimension of a linear subspace to the class of convex cones. The main technical result demonstrates that the sequence of conic intrinsic volumes of a convex cone concentrates sharply near the statistical dimension. This fact leads to an approximate version of the conic kinematic formula that gives bounds on the probability that a randomly oriented cone shares a ray with a fixed cone.

Item Type:MIMS Preprint
Uncontrolled Keywords:Compressed sensing, convex optimization, convex geometry, concentration of measure, phase transition
Subjects:MSC 2000 > 51 Geometry (See also algebraic geometry)
MSC 2000 > 52 Convex and discrete geometry
MSC 2000 > 60 Probability theory and stochastic processes
MSC 2000 > 65 Numerical analysis
MSC 2000 > 90 Operations research, mathematical programming
MIMS number:2013.14
Deposited By:Dr. Martin Lotz
Deposited On:28 March 2013

Download Statistics: last 4 weeks
Repository Staff Only: edit this item