Gaohang Yu, Zengxin Wei and Lutai Guan
		Key words:
		trust region method, nonconvex minimization, global convergence, superlinear convergence
		Mathematices Subject Classification: 90C30, 90C53, 65K05

ONLINE SUBSCRIPTION (Institutional Subscription Only) Copyright© 2006 Yokohama Publishers Back

---

	Abstract:
	By using a modified BFGS update, a globally convergent BFGS-type trust region method was proposed for unconstrained optimization problems in this paper. A favorable property of this method is that the trust region subproblem is always a strictly convex quadratic programming. Moreover, this property is independent of the convexity of the objective function. Under certain conditions, the superlinear convergence of this method was established. Preliminary numerical results have been presented, which show that the modified method is efficient.

A modified BFGS-trust region method for nonconvex minimization

---

Special Issue on Conjugate Gradient and Quasi-Newton Methods for Nonlinear Optimization
Volume 2, Number 1, January 2006, pp. 119-133