Learning Unknown Structure in CRFs via Adaptive Gradient Projection Method

Abstract

This paper focuses on learning unknown structure in conditional random fields (CRFs), especially learning both the structure and parameters of a CRF model simultaneously. By adding the l 2 -regularization to node parameters and the group l 1 -regularization to edge parameters, this structure learning problem can be cast as a convex minimization problem. Then an adaptive gradient method is proposed to solve the minimization problem. Extensive simulation experiments are presented to show the performance of the proposed approach for learning unknown structure.

A New Family of Hybrid Conjugate Gradient Methods for Unconstrained Optimization

“…where c is a positive constant. For more information on the theories and applications of CG methods, we refer the reader to [20,21,22,23,25,26,32]. Each CG method has very striking features that makes it adaptable to some sets of unconstrained problems.…”