We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Second-Order Optimality and Beyond: Characterization and Evaluation Complexity in Convexly Constrained Nonlinear Optimization.
- Authors
Cartis, Coralia; Gould, Nick I. M.; Toint, Philippe L.
- Abstract
High-order optimality conditions for convexly constrained nonlinear optimization problems are analysed. A corresponding (expensive) measure of criticality for arbitrary order is proposed and extended to define high-order ϵ<inline-graphic></inline-graphic>-approximate critical points. This new measure is then used within a conceptual trust-region algorithm to show that if derivatives of the objective function up to order q≥1<inline-graphic></inline-graphic> can be evaluated and are Lipschitz continuous, then this algorithm applied to the convexly constrained problem needs at most O(ϵ-(q+1))<inline-graphic></inline-graphic> evaluations of f and its derivatives to compute an ϵ<inline-graphic></inline-graphic>-approximate qth-order critical point. This provides the first evaluation complexity result for critical points of arbitrary order in nonlinear optimization. An example is discussed, showing that the obtained evaluation complexity bounds are essentially sharp.
- Subjects
NONLINEAR equations; COMPUTATIONAL complexity; MATHEMATICAL optimization; LIPSCHITZ spaces; NUMERICAL analysis
- Publication
Foundations of Computational Mathematics, 2018, Vol 18, Issue 5, p1073
- ISSN
1615-3375
- Publication type
Article
- DOI
10.1007/s10208-017-9363-y