Globally convergent limited memory bundle method for large-scale nonsmooth optimization

Many practical optimization problems involve nonsmooth (that is, not necessarily differentiable) functions of thousands of variables. In the paper [Haarala, Miettinen, M¨akel¨a, Optimization Methods and Software, 19, (2004), pp. 673–692] we have described an efficient method for large-scale nonsmooth optimization. In this paper, we introduce a new variant of this method and prove its global convergence for locally Lipschitz continuous objective functions, which are not necessarily differentiable or convex. In addition, we give some encouraging results from numerical experiments. Keywords Nondifferentiable programming · large-scale optimization · bundle methods · limited memory variable metric methods · global convergence

• Type: Journal
• Year: 2007
• Where: MP
• Authors: Napsu Haarala, Kaisa Miettinen, Marko M. Mäkelä
• Keywords: Global Convergence, Large-scale Nonsmooth Optimization, Many Practical Optimization, MP 2007