Abstract
AFSQP is a Sequential Quadratic Programming algorithm which obtains global convergence through an adaptive filter technique. This adaptivity is the major innovation in this work. The resulting algorithm can deal with constraints involving different length scales without requiring their normalization. The effort related to gradients computation is compensated by achieving superlinear local convergence rate (under some hypothesis on the problem, the algorithm can reach quadratic rates). Second order derivatives are approximated with classical BFGS formula and need not to be computed. We describe the theoretical background of the algorithm as well as its implementation details. A comparison between AFSQP and four different SQP implementations is performed considering several small and medium scale problems selected within Hoch and Schittkowski suite. We focus attention on the number of point evaluations required.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
NEOS Server, http://neos.mcs.anl.gov/neos/
Powell, M.J.D.: A fast algorithm for nonlinearly constrained optimization calculations. In: Watson, G.A. (ed.) Numerical Analysis, Dundee, pp. 144–157. Springer, Berlin (1977)
Schittkowski, K.: NLPQLP: a Fortran implementation of a sequential quadratic programming algorithm with distributed and non-monotone line search-user’s guide, Report, Department of Computer Science, University of Bayreuth (2006)
Gould, N.I.M., Toint, P.L.: SQP Methods for Large-Scale Nonlinear Programming. System Modelling and Optimization, 149–178 (1999)
Fletcher, R., Leyffer, S., Toint, P.L.: A Brief History of Filter Methods. SIAG/Optimization Views-and-News 18(1), 2–12 (2007)
Gould, N.I.M., Leyffer, S., Toint, P.L.: A Multidimensional Filter Algorithm for Nonlinear Equations and Nonlinear Least Squares. SIAM J. Optim. 15, 17–38 (2003)
Hock, W., Schittkowski, K.: Test Examples for Nonlinear Programming Codes. Lecture Notes in Economics and Mathematical Systems, vol. 187. Springer, Berlin (1981)
Fletcher, R., Leyffer, S.: Nonlinear programming without a penalty function. Mathematical Programming 91, 239–269 (2002)
Macconi, M., Morini, B., Porcelli, M.: Trust-region quadratic methods for nonlinear systems of mixed equalities and inequalities. Applied Numerical Mathematics (2008)
Nocedal, J., Wright, S.J.: Numerical Optimization. Springer, New York (1999)
Schittkowski, K.: NLPQLP: a Fortran implementation of a sequential quadratic programming algorithm with distributed and non-monotone line search-user’s guide, Report, Department of Computer Science, University of Bayreuth (2006)
Gill, P.E., Murray, W., Saunders, M.A.: SNOPT: An SQP Algorithm for Large-Scale Constrained Optimization. SIAM Review 47(1), 99–131
Shen, C., Xue, W., Pu, D.: Global convergence of a tri-dimensional filter SQP algorithm based on the line search method. Applied Numerical Mathematics 59, 235–250 (2009)
Dolan, E.D., Mor, J.: Benchmarking optimization software with performance profiles. Mathematical Programming 91, 201–213 (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Turco, A. (2010). Adaptive Filter SQP. In: Blum, C., Battiti, R. (eds) Learning and Intelligent Optimization. LION 2010. Lecture Notes in Computer Science, vol 6073. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-13800-3_6
Download citation
DOI: https://doi.org/10.1007/978-3-642-13800-3_6
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-13799-0
Online ISBN: 978-3-642-13800-3
eBook Packages: Computer ScienceComputer Science (R0)