Continuous OptimizationThe Dai–Liao nonlinear conjugate gradient method with optimal parameter choices
Introduction
Conjugate gradient (CG) methods comprise a class of unconstrained optimization algorithms characterized by low memory requirements and strong global convergence properties (Dai et al., 1999) which made them popular for engineers and mathematicians engaged in solving large-scale problems in the following form:where is a smooth nonlinear function and its gradient is available. The iterative formula of a CG method is given byin which is a steplength to be computed by a line search procedure and is the search direction defined bywhere and is a scalar called the CG (update) parameter.
The steplength is usually chosen to satisfy certain line search conditions (Sun & Yuan, 2006). Among them, the so-called strong Wolfe conditions (Wolfe, 1969) have attracted special attention in the convergence analyses and the implementations of CG methods, requiring thatwhere .
Different choices for the CG parameter lead to different CG methods (Hager & Zhang, 2006b). Based on an extended conjugacy condition, one of the essential CG methods has been proposed by Dai and Liao (2001) (DL), with the following CG parameter:where t is a nonnegative parameter and . Note that if , then reduces to the CG parameter proposed by Hestenes and Stiefel (1952). Also, the CG parameter proposed by Hager and Zhang (2005) (HZ), i.e.can be viewed as an adaptive version of (1.5) corresponding to , where denotes the Euclidean norm. Similarly, the CG parameter suggested by Dai and Kou (2013) (DK), i.e.in which is a parameter corresponding to the scaling factor in the scaled memoryless BFGS method (Sun & Yuan, 2006), can be considered as another adaptive version of (1.5) corresponding to . In (Dai & Liao, 2001), it has been shown that a CG method in the form of (1.1), (1.2) with is globally convergent for uniformly convex functions.
The approach of Dai and Liao has been paid special attention to by many researches. In several efforts, modified secant equations have been applied to make modifications on the DL method. For example, Yabe and Takano (2004) used the modified secant equation proposed by Zhang, Deng, and Chen (1999). Also, Zhou and Zhang (2006) applied the modified secant equation proposed by Li and Fukushima (2001). Li, Tang, and Wei (2007) used the modified secant equation proposed by Wei, Li, and Qi (2006). Ford, Narushima, and Yabe (2008) employed the multi-step quasi-Newton equations proposed by Ford and Moghrabi (1994). Babaie-Kafaki, Ghanbari, and Mahdavi-Amiri (2010) applied a revised form of the modified secant equation proposed by Zhang et al. (1999) and the modified secant equation proposed by Yuan (1991). Furthermore, in several other attempts, the modified versions of suggested in (Babaie-Kafaki et al., 2010, Ford et al., 2008, Li et al., 2007, Yabe and Takano, 2004, Zhou and Zhang, 2006) have been used to achieve descent CG methods. Examples include the studies made by Narushima and Yabe, 2012, Sugiki et al., 2012, Livieris and Pintelas, 2012.
Here, based on a singular value study on the DL method, two nonlinear CG methods are proposed. The remainder of this work is organized as follows. In Section 2, the methods are suggested and their global convergence analysis is discussed. In Section 3, they are numerically compared with the CG methods proposed by Hager and Zhang, and Dai and Kou, and comparative testing results are reported. Finally, conclusions are made in Section 4.
Section snippets
Two modified nonlinear conjugate gradient methods
Based on Perry’s point of view (Perry, 1976), it is notable that from (1.2), (1.5), search directions of the DL method can be written as:whereSo, the DL method can be considered as a quasi-Newton method in which the inverse Hessian is approximated by the nonsymmetric matrix . Since presents a rank-two update, its determinant can be computed by (Sun & Yuan, 2006, chap. 1)Hence, if and the line search
Numerical experiments
Here, we present some numerical results obtained by applying C++ implementations of the CG methods in the form of (1.1), (1.2) in which defined by (1.5) with the suggested two optimal choices defined by (2.17) and defined by (2.20), here respectively called M1 and M2, the HZ method with the CG parameter (1.6), and the DK method with the following optimal choice for in (1.7), suggested by Dai and Kou (2013):The results are extended by further study on the
Conclusions
Based on a singular value study on the matrix which generates the search directions of the Dai–Liao nonlinear conjugate gradient method, two modified conjugate gradient methods have been suggested. Global convergence of the methods has been briefly discussed. Numerical comparisons have been made between the implementations of the proposed methods, and the conjugate gradient methods proposed by Hager and Zhang, and Dai and Kou, on a set of 145 unconstrained optimization test problems of the
Acknowledgements
This research was supported by the Research Councils of Semnan University and Ferdowsi University of Mashhad. The authors are grateful to Professor William W. Hager for providing the test problems and the CG_Descent code freely. They also thank the two anonymous reviewers for their valuable comments and suggestions helped to improve the quality of this work.
References (27)
- et al.
Two new conjugate gradient methods based on modified secant equations
Journal of Computational and Applied Mathematics
(2010) - et al.
Multi-step quasi-Newton methods for optimization
Journal of Computational and Applied Mathematics
(1994) - et al.
A modified BFGS method and its global convergence in nonconvex minimization
Journal of Computational and Applied Mathematics
(2001) - et al.
New conjugacy condition and related new conjugate gradient methods for unconstrained optimization
Journal of Computational and Applied Mathematics
(2007) - et al.
Conjugate gradient methods based on secant conditions that generate descent search directions for unconstrained optimization
Journal of Computational and Applied Mathematics
(2012) - et al.
New quasi-Newton methods for unconstrained optimization problems
Applied Mathematics and Computation
(2006) Open problems in conjugate gradient algorithms for unconstrained optimization
Bulletin of the Malaysian Mathematical Sciences Society
(2011)- et al.
Convergence properties of nonlinear conjugate gradient methods
SIAM Journal on Optimization
(1999) - et al.
A nonlinear conjugate gradient algorithm with an optimal property and an improved Wolfe line search
SIAM Journal on Optimization
(2013) - et al.
New conjugacy conditions and related nonlinear conjugate gradient methods
Applied Mathematics and Optimization
(2001)
Benchmarking optimization software with performance profiles
Mathematical Programming
Multi-step nonlinear conjugate gradient methods for unconstrained minimization
Computational Optimization and Applications
CUTEr: A constrained and unconstrained testing environment, revisited
ACM Transactions on Mathematical Software
Cited by (83)
An accelerated descent CG algorithm with clustering the eigenvalues for large-scale nonconvex unconstrained optimization and its application in image restoration problems
2024, Journal of Computational and Applied MathematicsTwo families of self-adjusting spectral hybrid DL conjugate gradient methods and applications in image denoising
2023, Applied Mathematical ModellingModified optimal Perry conjugate gradient method for solving system of monotone equations with applications
2023, Applied Numerical MathematicsA new CG algorithm based on a scaled memoryless BFGS update with adaptive search strategy, and its application to large-scale unconstrained optimization problems
2021, Journal of Computational and Applied MathematicsOn a Scaled Symmetric Dai–Liao-Type Scheme for Constrained System of Nonlinear Equations with Applications
2024, Journal of Optimization Theory and Applications