An Efficient Penalty Method without a Line Search for Nonlinear Optimization
Abstract
:1. Introduction
2. The Problem
3. Formulation of the Perturbed Problem of (P)
3.1. Existence and Uniqueness of Optimal Solution
3.2. Convergence of the Solution
4. Computational Resolution of the Perturbed Problem
4.1. The Descent Direction
4.2. Computation of the Displacement Step
4.2.1. Line Search Methods
4.2.2. Approximate Functions Techniques
4.2.3. New Approximate Functions Approach
4.3. Minimize an Auxiliary Function
- If we obtain
- If we obtain
- If we have
- If is the only root of the second-degree equation that belongs to the domain of definition of We obtain Both roots are
4.4. The Objective Function f Is
4.4.1. Linear
4.4.2. Convex
- Where We have the following:
- Where The computation of is performed through a dichotomous procedure (see Remark 3).
5. Description of the Algorithm and Numerical Simulations
5.1. Description of the Algorithm
- Start with
- Calculate d and
- If calculate , and
- Determine following (8), (10), or (9) depending on the linear or nonlinear case.
- Take the new iterate and go back to step 2.
- If a well approximation of has been obtained.
- (a)
- If and return to step 2.
- (b)
- If STOP: a well approximate solution of has been obtained.
5.2. Numerical Simulations
5.2.1. Examples with a Fixed Size
Nonlinear Convex Objective
Example | st1 | st2 | LS | ||||||
iter | Time (s) | iter | Time (s) | iter | Time (s) | ||||
1 | 12 | 0.0006 | 19 | 0.0015 | 6 | 0.0091 | |||
2 | 5 | 0.0004 | 9 | 0.0009 | 44 | 0.099 | |||
3 | 3 | 0.0001 | 5 | 0.0006 | 65 | 0.89 |
5.2.2. Example with a Variable Size
The Objective Function f Is
Size | st1 | st2 | LS | |||||
iter | Time (s) | iter | Time (s) | iter | Time (s) | |||
1 | 0.0021 | 2 | 0.0039 | 9 | 0.0512 | |||
1 | 0.0031 | 3 | 0.0045 | 13 | 0.0821 | |||
2 | 0.0049 | 3 | 0.0032 | 17 | 0.3219 | |||
2 | 0.0053 | 4 | 0.0088 | 19 | 0.5383 | |||
2 | 0.0088 | 4 | 0.0098 | 22 | 0.9220 | |||
3 | 0.0096 | 5 | 0.0125 | 26 | 9.2647 |
ex() | st1 | st2 | LS | ||||
iter | Time (s) | iter | Time (s) | iter | Time (s) | ||
5 | 0.9968 | 4 | 0.9699 | 26 | 19.5241 | ||
7 | 18.1448 | 5 | 9.6012 | 35 | 86.1259 | ||
12 | 36.3259 | 5 | 19.0099 | 23 | 98.2354 | ||
21 | 56.9912 | 17 | 41.1012 | 33 | 109.2553 | ||
28 | 140.1325 | 23 | 95.6903 | 40 | 1599.1596 |
ex() | st1 | st2 | LS | ||||
iter | Time (s) | iter | Time (s) | iter | Time (s) | ||
1 | 0.0001 | 2 | 0.0012 | 4 | 0.0236 | ||
2 | 0.0021 | 3 | 0.0033 | 5 | 0.7996 | ||
2 | 0.0043 | 3 | 0.0201 | 5 | 1.5289 | ||
2 | 3.0901 | 4 | 5.9619 | 12 | 22.1254 |
6. Conclusions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Frank, M.; Wolfe, B. An algorithm for quadratic programming. Nav. Res. Logist. Q. 1956, 3, 95–110. [Google Scholar] [CrossRef]
- Wolfe, P. A Duality Theorem for Nonlinear Programming. Q. Appl. Math. 1961, 19, 239–244. [Google Scholar] [CrossRef]
- Bracken, J.; M1cCormiek, G.P. Selected Applications of Nonlinear Programming; John Wiley & Sons, Inc.: New York, NY, USA, 1968. [Google Scholar]
- Fiacco, A.V.; McCormick, G.P. Nonlinear Programming: Sequential Unconstrained Minimization Techniques; John Wiley & Sons, Inc.: New York, NY, USA, 1968. [Google Scholar]
- Bonnans, J.-F.; Gilbert, J.-C.; Lemaréchal, C.; Sagastizàbal, C. Numerical Optimization: Theoretical and Practical Aspects; Mathematics and Applications; Springer: Berlin/Heidelberg, Germany, 2003; Volume 27. [Google Scholar]
- Evtushenko, Y.G.; Zhadan, V.G. Stable barrier-projection and barrier-Newton methods in nonlinear programming. In Optimization Methods and Software; Taylor & Francis: Abingdon, UK, 1994; Volume 3, pp. 237–256. [Google Scholar]
- Nestrov, Y.E.; Nemiroveskii, A. Interior-Point Polynomial Algorithms in Convex Programming; SIAM: Philadelphia, PA, USA, 1994. [Google Scholar]
- Wright, S.J. Primal–Dual Interior Point Methods; SIAM: Philadelphia, PA, USA, 1997. [Google Scholar]
- Ye, Y. Interior Point Algorithms: Theory and Analysis. In Discrete Mathematics Optimization; Wiley-Interscience Series; John Wiley & Sons: New York, NY, USA, 1997. [Google Scholar]
- Powell, M.J.D. Karmarkar’s Algorithm: A View from Nonlinear Programming; Department of Applied Mathematics and Theoretical Physics, University of Cambridge: Cambridge, UK, 1989; Volume 53. [Google Scholar]
- Rosen, J.B. The Gradient Projection Method for Nonlinear Programming. Soc. Ind. Appl. Math. J. Appl. Math. 1960, 8, 181–217. [Google Scholar] [CrossRef]
- Rosen, J.B. The Gradient Projection Method for Nonlinear Programming. Soc. Ind. Appl. Math. J. Appl. Math. 1961, 9, 514–553. [Google Scholar] [CrossRef]
- Ouriemchi, M. Résolution de Problèmes non Linéaires par les Méthodes de Points Intérieurs. Théorie et Algorithmes. Doctoral Thesis, Université du Havre, Havre, France, 2006. [Google Scholar]
- Forsgren, A.; Gill, P.E.; Wright, M.H. Interior Methods for Nonlinear Optimization; SIAM: Philadelphia, PA, USA, 2002; Volume 44, pp. 525–597. [Google Scholar]
- Crouzeix, J.P.; Merikhi, B. A logarithm barrier method for semidefinite programming. RAIRO-Oper. Res. 2008, 42, 123–139. [Google Scholar] [CrossRef]
- Menniche, L.; Benterki, D. A Logarithmic Barrier Approach for Linear Programming. J. Computat. Appl. Math. 2017, 312, 267–275. [Google Scholar] [CrossRef]
- Cherif, L.B.; Merikhi, B. A Penalty Method for Nonlinear Programming. RAIRO-Oper. Res. 2019, 53, 29–38. [Google Scholar] [CrossRef]
- Chaghoub, S.; Benterki, D. A Logarithmic Barrier Method Based on a New Majorant Function for Convex Quadratic Programming. IAENG Int. J. Appl. Math. 2021, 51, 563–568. [Google Scholar]
- Leulmi, A. Etude d’une Méthode Barrière Logarithmique via Minorants Functions pour la Programmation Semi-Définie. Doctoral Thesis, Université de Biskra, Biskra, Algeria, 2018. [Google Scholar]
- Leulmi, A.; Merikhi, B.; Benterki, D. Study of a Logarithmic Barrier Approach for Linear Semidefinite Programming. J. Sib. Fed. Univ. Math. Phys. 2018, 11, 300–312. [Google Scholar]
- Leulmi, A.; Leulmi, S. Logarithmic Barrier Method via Minorant Function for Linear Programming. J. Sib. Fed. Univ. Math. Phys. 2019, 12, 191–201. [Google Scholar] [CrossRef]
- Wolkowicz, H.; Styan, G.P.H. Bounds for Eigenvalues Using Traces. Lin. Alg. Appl. 1980, 29, 471–506. [Google Scholar] [CrossRef]
- Crouzeix, J.-P.; Seeger, A. New bounds for the extreme values of a finite sample of real numbers. J. Math. Anal. Appl. 1996, 197, 411–426. [Google Scholar] [CrossRef]
- Bazraa, M.S.; Sherali, H.D.; Shetty, C.M. Nonlinear Programming, Willey-Interscience; John Wiley & Sons, Inc.: Hoboken, NJ, USA; Toronto, ON, Canada, 2006. [Google Scholar]
- Shannon, E. A mathematical theory of communication. Bell Syst. Tech. J. 1948, 27, 379–423+623–656. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Leulmi, A. An Efficient Penalty Method without a Line Search for Nonlinear Optimization. Axioms 2024, 13, 176. https://doi.org/10.3390/axioms13030176
Leulmi A. An Efficient Penalty Method without a Line Search for Nonlinear Optimization. Axioms. 2024; 13(3):176. https://doi.org/10.3390/axioms13030176
Chicago/Turabian StyleLeulmi, Assma. 2024. "An Efficient Penalty Method without a Line Search for Nonlinear Optimization" Axioms 13, no. 3: 176. https://doi.org/10.3390/axioms13030176
APA StyleLeulmi, A. (2024). An Efficient Penalty Method without a Line Search for Nonlinear Optimization. Axioms, 13(3), 176. https://doi.org/10.3390/axioms13030176