- Open Access
Error bounds of regularized gap functions for weak vector variational inequality problems
Journal of Inequalities and Applications volume 2014, Article number: 331 (2014)
In this paper, by the nonlinear scalarization method, a global error bound of a weak vector variational inequality is established via a regularized gap function. The result extends some existing results in the literature.
Throughout this paper, let K be a closed convex subset of an Euclidean space and be a continuously differentiable mapping. We consider a weak vector variational inequality (WVVI) of finding such that
where is a closed convex and pointed cone with nonempty interior intC. (WVVI) was firstly introduced by Giannessi . It has been shown to have many applications in vector optimization problems and traffic equilibrium problems (e.g., [2, 3]).
Error bounds are to depict the distance from a feasible solution to the solution set, and have played an important role not only in sensitivity analysis but also in convergence analysis of iterative algorithms. Recently, kinds of error bounds have been presented for weak vector variational inequalities in [4–7]. By using a scalarization approach of Konnov , Li and Mastroeni  established the error bounds for two kinds of (WVVIs) with set-valued mappings. By a regularized gap function and a D-gap function for a weak vector variational inequality, Charitha and Dutta  obtained the error bounds of (WVVI), respectively. Recently, in virtue of the regularized gap functions, Sun and Chai  studied some error bounds for generalized (WVVIs). By using the image space analysis, Xu and Li  got a gap function for (WVVI). Then, they established an error bound for (WVVI) without the convexity of the constraint set. These papers have a common characteristic: the solution set of (WVVI) is a singleton [6, 7]. Even though the solution set of (WVVI) is not a singleton [4, 5], the solution set of the corresponding variational inequality (VI) is a singleton, when their results reduce to (VI).
In this paper, by the nonlinear scalarization method, we study a global error bound of (WVVI). This paper is organized as follows. In Section 2, we establish a global error bound of (VI) via the generalized gap functions. In Section 3, we discuss a global error bound of (WVVI) by the nonlinear scalarization method.
2 A global error bound of (VI)
Let be a proper lower semicontinuous function, and let . h has a global error bound if there exists such that
where and if S is nonempty and if S is empty. is said to be coercive on K if
is said to be strongly monotone on with the modulus if
In this section, we establish a global error bound of (VI) of finding such that
where is a continuously differentiable mapping.
To study the error bound of (VI), we need to construct a class of merit functions which were made to reformulate (VI) as an optimization problem; see [9–16]. One of such functions is a generalized regularized gap function  defined by
where is a real-valued function with the following properties:
(P1) φ is continuously differentiable on .
(P2) , and the equality holds if and only if .
(P3) is uniformly strongly convex on with the modulus in the sense that
where denotes the partial derivative of φ with respect to the second variable.
(P4) is uniformly Lipschitz continuous on with the modulus α, i.e., for all ,
Now we recall some properties of φ in (1).
Proposition 2.1 The following statements hold for each :
if and only if .
Remark 2.1 In light of (ii) in Proposition 2.1, it holds true that .
Then we list some basic properties of the generalized regularized gap function .
Proposition 2.2 The following conclusions are valid for (VI).
For every , there exists a unique vector at which the infimum in (1) is attained, i.e.,
is a gap function of (VI).
if and only if x is a solution of (VI).
is continuously differentiable on with
and are both locally Lipschitz on .
If f is coercive on K, then (VI) has a nonempty compact solution set.
It follows from (ii) and (iii) that we only need to prove (vii) for . Since is the minimizer of the function
the first-order optimality condition implies that
Letting , we get
It follows from (P3) that the mapping G is strongly convex on with the modulus , i.e.,
Letting and , by , we obtain
Thus, one has . □
Theorem 2.1 Let f be coercive on K and . Assume that φ satisfies
(P5) , .
Suppose further that the following condition holds:
where S is the solution set of (VI). Then has a global error bound with the modulus
Proof It follows from (vi) of Proposition 2.2 that S is a nonempty compact set of K. If , then the assertion obviously holds. Let . Then . For brevity, we denote and . It follows from [, Theorem 2.5] that we only need to prove
It follows from (iv) of Proposition 2.2 that
By (P5) and (2), we have
It follows from (iii) of Proposition 2.1 that
In light of (4) and (vii) of Proposition 2.2, we have
If , then it follows from that
If , then
Hence, (3) follows from (5) and (6). The proof is complete. □
Now we use two examples to show that (2) cannot be dropped and that Theorem 2.1 is applicable, respectively.
Example 2.1 Consider , , and . Then we can easily get that , , and . It is clear that f is coercive on K and . Thus, (2) does not hold. Moreover, it is obvious that does not have a global error bound.
Example 2.2 Consider , , and . Then we can easily get that , , , and . It is clear that f is coercive on K and (2) holds. Thus, it follows from Theorem 2.1 that has a global error bound.
Corollary 2.1 Let f be strongly monotone on with the modulus and . Assume that φ satisfies (P5). Then has a global error bound with the modulus
Proof Let and . Since f is continuously differentiable, then
where as . Since f is strongly monotone with the modulus λ, one has
which implies that
Thus, (2) holds. Moreover, the strong monotonicity of f implies the coerciveness of f (cf. [, Remark 2.1]). Thus, by Theorem 2.1, we get that has a global error bound. □
3 A global error bound of (WVVI)
In this section, by the nonlinear scalarization method and by Theorem 2.1, we discuss a global error bound of (WVVI). The dual cone of C is defined by . For each , , where denotes the value of ξ at z. Let and . It is well known that is a compact convex base of .
Lemma 3.1 
where and S is the solution set of (WVVI).
Recall the generalized regularized gap function for (WVVI) which is defined by
where . When , the generalized regularized gap function reduces to the regularized gap function which was defined in .
Theorem 3.1 Let . Assume that φ satisfies (P5). For each , suppose that is coercive on K, and that the following condition holds:
Then has a global error bound with the modulus
Proof It follows from (vi) of Proposition 2.2 that is a nonempty compact set of K for each . If , then the assertion obviously holds. Let . Then and there exists such that . It follows from Theorem 2.1 that
where . Thus, by Lemma 3.1, one has
Hence, has a global error bound with the modulus . □
Remark 3.1 If is strongly monotone with the modulus for and , it follows from [, Proposition 2.3] that
Moreover, the strong monotonicity of implies the coerciveness of (cf. [, Remark 2.1]) and that (VI) has a unique solution (cf. [, Theorem 2.3.3]). Thus, by Theorem 3.1, we get that has a global error bound. Hence, our results extend those of [, Theorem 2.9].
Giannessi F: Theorems of the alternative, quadratic programs and complementarity problems. In Variational Inequalities and Complementarity Problems. Edited by: Cottle RW, Giannessi F, Lions JL. Wiley, New York; 1980:151–186.
Chen GY, Goh CJ, Yang XQ: Vector network equilibrium problems and nonlinear scalarization methods. Math. Methods Oper. Res. 1999, 49: 239–253.
Li SJ, Teo KL, Yang XQ: A remark on a standard and linear vector network equilibrium problem with capacity constraints. Eur. J. Oper. Res. 2008, 184: 13–23. 10.1016/j.ejor.2005.11.059
Charitha C, Dutta J: Regularized gap functions and error bounds for vector variational inequalities. Pac. J. Optim. 2010, 6: 497–510.
Li J, Mastroeni G: Vector variational inequalities involving set-valued mappings via scalarization with applications to error bounds for gap functions. J. Optim. Theory Appl. 2010, 145: 355–372. 10.1007/s10957-009-9625-1
Sun XK, Chai Y: Gap functions and error bounds for generalized vector variational inequalities. Optim. Lett. 2014, 8: 1663–1673. 10.1007/s11590-013-0685-7
Xu YD, Li SJ: Gap functions and error bounds for weak vector variational inequalities. Optimization 2014, 63: 1339–1352. 10.1080/02331934.2012.721115
Konnov IV: A scalarization approach for vector variational inequalities with applications. J. Glob. Optim. 2005, 32: 517–527. 10.1007/s10898-003-2688-x
Auchmuty G: Variational principles for variational inequalities. Numer. Funct. Anal. Optim. 1989, 10: 863–874. 10.1080/01630568908816335
Auslender A: Optimization, Méthodes Numériques. Masson, Paris; 1976.
Facchinei F, Pang JS: Finite-Dimensional Variational Inequalities and Complementarity Problems. Springer, Berlin; 2003.
Fukushima M: Equivalent differentiable optimization problems and descent methods for asymmetric variational inequality problems. Math. Program. 1992, 53: 99–110. 10.1007/BF01585696
Harker PT, Pang JS: Finite-dimensional variational inequality and nonlinear complementarity problems: a survey of theory, algorithms, and applications. Math. Program. 1990, 48: 161–220. 10.1007/BF01582255
Huang LR, Ng KF: Equivalent optimization formulations and error bounds for variational inequality problems. J. Optim. Theory Appl. 2005, 125: 299–314. 10.1007/s10957-004-1839-7
Ng KF, Tan LL: Error bounds of regularized gap functions for nonsmooth variational inequality problems. Math. Program. 2007, 110: 405–429. 10.1007/s10107-006-0007-2
Wu JH, Florian M, Marcotte P: A general descent framework for the monotone variational inequality problem. Math. Program. 1993, 61: 281–300. 10.1007/BF01582152
Yamashita N, Taji K, Fukushima M: Unconstrained optimization reformulations of variational inequality problems. J. Optim. Theory Appl. 1997, 92: 439–456. 10.1023/A:1022660704427
Li G, Tang C, Wei Z: Error bound results for generalized D-gap functions of nonsmooth variational inequality problems. J. Comput. Appl. Math. 2010, 233: 2795–2806. 10.1016/j.cam.2009.11.025
Ng KF, Zheng XY: Error bounds for lower semicontinuous functions in normed spaces. SIAM J. Optim. 2001, 12: 1–17. 10.1137/S1052623499358884
Lee GM, Kim DS, Lee BS, Yen ND: Vector variational inequality as a tool for studying vector optimization problems. Nonlinear Anal. 1998, 34: 745–765. 10.1016/S0362-546X(97)00578-6
Jiang HY, Qi LQ: Local uniqueness and convergence of iterative methods for nonsmooth variational inequalities. J. Math. Anal. Appl. 1995, 196: 314–331. 10.1006/jmaa.1995.1412
Li G, Ng KF: Error bounds of generalized D-gap functions for nonsmooth and nonmonotone variational inequality problems. SIAM J. Optim. 2009, 20: 667–690. 10.1137/070696283
This research was supported by the Natural Science Foundation of Shaanxi Province, China (Grant number: 2014JQ1023).
The author declares that he has no competing interests.
About this article
Cite this article
Li, M. Error bounds of regularized gap functions for weak vector variational inequality problems. J Inequal Appl 2014, 331 (2014). https://doi.org/10.1186/1029-242X-2014-331
- error bound
- regularized gap function
- weak vector variational inequality