An Efficient Re-Scaled Perceptron Algorithm for Conic Systems
The classical perceptron algorithm is an elementary row-action/relaxation algorithm for solving a homogeneous linear inequality system Ax > 0. A natural condition measure associated with this algorithm is the Euclidean width T of the cone of feasible solutions, and the iteration complexity of the perceptron algorithm is bounded by 1/T^2, see Rosenblatt 1962. Dunagan and Vempala have developed a re-scaled version of the perceptron algorithm with an improved complexity of O(n ln(1/T)) iterations (with high probability), which is theoretically efficient in T, and in particular is polynomial-time in the bit-length model. We explore extensions of the concepts of these perceptron methods to the general homogeneous conic system Ax is an element of a set int K where K is a regular convex cone. We provide a conic extension of the re-scaled perceptron algorithm based on the notion of a deep-separation oracle of a cone, which essentially computes a certificate of strong separation. We give a general condition under which the re-scaled perceptron algorithm is itself theoretically efficient; this includes the cases when K is the cross-product of half-spaces, second-order cones, and the positive semi-definite cone.
|Date of creation:||27 Apr 2007|
|Contact details of provider:|| Postal: MASSACHUSETTS INSTITUTE OF TECHNOLOGY (MIT), SLOAN SCHOOL OF MANAGEMENT, 50 MEMORIAL DRIVE CAMBRIDGE MASSACHUSETTS 02142 USA|
Web page: http://mitsloan.mit.edu/
More information through EDIRC
|Order Information:|| Postal: MASSACHUSETTS INSTITUTE OF TECHNOLOGY (MIT), SLOAN SCHOOL OF MANAGEMENT, 50 MEMORIAL DRIVE CAMBRIDGE MASSACHUSETTS 02142 USA|
When requesting a correction, please mention this item's handle: RePEc:mit:sloanp:37304. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Christian Zimmermann)
If references are entirely missing, you can add them using this form.