Nonconvex Case Research Articles

Group Update Method for Sparse Minimax Problems

A group update algorithm is presented for solving minimax problems with a finite number of functions, whose Hessians are sparse. The method uses the gradient evaluations as efficiently as possible by updating successively the elements in partitioning groups of the columns of every Hessian in the process of iterations. The chosen direction is determined directly by the nonzero elements of the Hessians in terms of partitioning groups. The local $$q$$q-superlinear convergence of the method is proved, without requiring the imposition of a strict complementarity condition, and the $$r$$r-convergence rate is estimated. Furthermore, two efficient methods handling nonconvex case are given. The global convergence of one method is proved, and the local $$q$$q-superlinear convergence and $$r$$r-convergence rate of another method are also proved or estimated by a novel technique. The robustness and efficiency of the algorithms are verified by numerical tests.

Recovery of Sparse Signal and Nonconvex Minimization

In this paper, we present sufficient conditions in term of the restricted isometry property (RIP) to guarantee perfect recovery of sparse signal in the noiseless case and stable recovery in the noisy case via Lq-minimization, especially for nonconvex case 0<q<1. Using RIP condition, we present sufficient conditions to guarantee perfect recovery of sparse signal in the noiseless case and stable recovery in the noisy case via Lq-minimization.

A penalty-based aggregation operator for non-convex intervals

In the case of real-valued inputs, averaging aggregation functions have been studied extensively with results arising in fields including probability and statistics, fuzzy decision-making, and various sciences. Although much of the behavior of aggregation functions when combining standard fuzzy membership values is well established, extensions to interval-valued fuzzy sets, hesitant fuzzy sets, and other new domains pose a number of difficulties. The aggregation of non-convex or discontinuous intervals is usually approached in line with the extension principle, i.e. by aggregating all real-valued input vectors lying within the interval boundaries and taking the union as the final output. Although this is consistent with the aggregation of convex interval inputs, in the non-convex case such operators are not idempotent and may result in outputs which do not faithfully summarize or represent the set of inputs. After giving an overview of the treatment of non-convex intervals and their associated interpretations, we propose a novel extension of the arithmetic mean based on penalty functions that provides a representative output and satisfies idempotency.

Characterizations of efficiency in vector optimization with $$\mathcal {C}(T)$$ C ( T ) -valued mappings

In this paper, under the assumption of pseudoconvexity, we present some characterizations of efficient solutions and weak efficient solutions for vector optimization problems with \(\mathcal {C}(T)\)-valued mappings by means of Clarke derivatives and subdifferentials. Our main results generalize the corresponding results in Winkler (SIAM J. Optim. 19:756–765, 2008) to nonconvex case.

Modified interactive chebyshev algorithm (MICA) for non-convex multiobjective programming

In this paper, I carry out an extension of the MICA method (modified interactive chebyshev algorithm) for non-convex multiobjective programming. This method is based on the Tchebychev method and in the reference point approach. At each iteration, the decision maker (DM) can provide aspiration levels (desirable values for the objective functions) and also, if the DM wishes, reservation levels (level under which the objective function is not considered acceptable). On the basis of this preferential information, a region of the nondominated objective set is defined. In the convex case, considering the aspiration vector as a reference point in an achievement scalarizing function and taking a set of weight vectors, the efficient solutions generated satisfy the reservation levels. In this work, I analyze the non-convex case. The main result of MICA is verified and demonstrated for the non-convex bi-objective case. The MICA method is not verified in general for multiobjective problems with three or more objective functions, which is demonstrated with a counterexample.

On the relaxation of variational integrals in metric Sobolev spaces

Abstract We give an extension of the theory of relaxation of variational integrals in classical Sobolev spaces to the setting of metric Sobolev spaces. More precisely, we establish a general framework to deal with the problem of finding an integral representation for “relaxed” variational functionals of variational integrals of the calculus of variations in the setting of metric measure spaces. We prove integral representation theorems, both in the convex and non-convex case, which extend and complete previous results in the setting of euclidean measure spaces to the setting of metric measure spaces. We also show that these integral representation theorems can be applied in the setting of Cheeger–Keith's differentiable structure.

On subdifferentials of a minimal time function in Hausdorff topological vector spaces

In a general Hausdorff topological vector space , we study the minimal time function associated with a nonempty closed (both cases: convex and nonconvex) set and a bounded closed convex set defined by . We prove and extend various important properties on directional derivatives and subdifferentials of at points in . These results are used to prove various new characterizations of the convex tangent cone, the convex normal cone, the Clarke tangent cone, the Bouligand tangent cone, and Clarke normal cone to in terms of the minimal time function for points inside , in Hausdorff topological vector spaces. Those characterizations are used to scalarize the tangential regularity of sets, in Hausdorff topological vector spaces, as the directional regularity of . Our results extend various existing results, in convex and nonconvex cases, from Banach spaces and normed vector spaces to Hausdorff topological vector spaces. Even in Banach spaces and in normed spaces our results are new, and they extend various existing results on the distance function to closed sets by taking to be the closed unit ball. Applications to normal and subidfferential regularities in normed vector spaces are also given in the last section of the paper.

Optimality Conditions for Quasi-Solutions of Vector Optimization Problems

In this paper, we deal with quasi-solutions of constrained vector optimization problems. These solutions are a kind of approximate minimal solutions and they are motivated by the Ekeland variational principle. We introduce several notions of quasi-minimality based on free disposal sets and we characterize these solutions through scalarization and Lagrange multiplier rules. When the problem fulfills certain convexity assumptions, these results are obtained by using linear separation and the Fenchel subdifferential. In the nonconvex case, they are stated by using the so-called Gerstewitz (Tammer) nonlinear separation functional and the Mordukhovich subdifferential.

Stochastic Optimization over a Pareto Set Associated with a Stochastic Multi-Objective Optimization Problem

We deal with the problem of minimizing the expectation of a real valued random function over the weakly Pareto or Pareto set associated with a Stochastic Multi-objective Optimization Problem, whose objectives are expectations of random functions. Assuming that the closed form of these expectations is difficult to obtain, we apply the Sample Average Approximation method in order to approach this problem. We prove that the Hausdorff---Pompeiu distance between the weakly Pareto sets associated with the Sample Average Approximation problem and the true weakly Pareto set converges to zero almost surely as the sample size goes to infinity, assuming that our Stochastic Multi-objective Optimization Problem is strictly convex. Then we show that every cluster point of any sequence of optimal solutions of the Sample Average Approximation problems is almost surely a true optimal solution. To handle also the non-convex case, we assume that the real objective to be minimized over the Pareto set depends on the expectations of the objectives of the Stochastic Optimization Problem, i.e. we optimize over the image space of the Stochastic Optimization Problem. Then, without any convexity hypothesis, we obtain the same type of results for the Pareto sets in the image spaces. Thus we show that the sequence of optimal values of the Sample Average Approximation problems converges almost surely to the true optimal value as the sample size goes to infinity.

A Bounded and Discretized Nelder-Mead Algorithm Suitable for RFIC Calibration

This paper describes a calibration technique for noisy and nonconvex circuit responses based on the Nelder-Mead direct search algorithm. As Nelder-Mead is intended for unconstrained optimization problems, we present an implementation of the algorithm which is suitable for bounded and discretized RFIC calibration problems. We apply the proposed algorithm to the problem of spurious tone reduction via VCO control line ripple minimization for a PLL operating at a frequency of 12 GHz. For this nonconvex calibration test case, we show that a gradient descent-based algorithm has difficulty in reducing the VCO control line ripple, while the proposed algorithm reduces the relative power of the first harmonic reference spurs by at least 10 dBc and effectively enables design complexity reduction in the supporting analog calibration circuitry.

Restricted Normal Cones and the Method of Alternating Projections: Applications

The method of alternating projections (MAP) is a common method for solving feasibility problems. While employed traditionally to subspaces or to convex sets, little was known about the behavior of the MAP in the nonconvex case until 2009, when Lewis, Luke, and Malick derived local linear convergence results provided that a condition involving normal cones holds and at least one of the sets is superregular (a property less restrictive than convexity). However, their results failed to capture very simple classical convex instances such as two lines in a three-dimensional space. In this paper, we extend and develop the Lewis-Luke-Malick framework so that not only any two linear subspaces but also any two closed convex sets whose relative interiors meet are covered. We also allow for sets that are more structured such as unions of convex sets. The key tool required is the restricted normal cone, which is a generalization of the classical Mordukhovich normal cone. Numerous examples are provided to illustrate the theory.

An Exact Algorithm for Nonconvex Quadratic Integer Minimization Using Ellipsoidal Relaxations

We propose a branch-and-bound algorithm for minimizing a not necessarily convex quadratic function over integer variables. The algorithm is based on lower bounds computed as continuous minima of the objective function over appropriate ellipsoids. In the nonconvex case, we use ellipsoids enclosing the feasible region of the problem. In spite of the nonconvexity, these minima can be computed quickly; the corresponding optimization problems are equivalent to trust-region subproblems. We present several ideas that allow us to accelerate the solution of the continuous relaxation within a branch-and-bound scheme and examine the performance of the overall algorithm by computational experiments. Good computational performance is shown especially for ternary instances.

Fractional order Riemann-Liouville integral inclusions with two independent variables and multiple delay

In the present paper we investigate the existence of solutions for a system of integral inclusions of fractional order with multiple delay. Our results are obtained upon suitable fixed point theorems, namely the Bohnenblust-Karlin fixed point theorem for the convex case and the Covitz-Nadler for the nonconvex case.

Gradient methods for minimizing composite functions

In this paper we analyze several new methods for solving optimization problems with the objective function formed as a sum of two terms: one is smooth and given by a black-box oracle, and another is a simple general convex function with known structure. Despite the absence of good properties of the sum, such problems, both in convex and nonconvex cases, can be solved with efficiency typical for the first part of the objective. For convex problems of the above structure, we consider primal and dual variants of the gradient method (with convergence rate $$O\left({1 \over k}\right)$$ ), and an accelerated multistep version with convergence rate $$O\left({1 \over k^2}\right)$$ , where $$k$$ is the iteration counter. For nonconvex problems with this structure, we prove convergence to a point from which there is no descent direction. In contrast, we show that for general nonsmooth, nonconvex problems, even resolving the question of whether a descent direction exists from a point is NP-hard. For all methods, we suggest some efficient “line search” procedures and show that the additional computational work necessary for estimating the unknown problem class parameters can only multiply the complexity of each iteration by a small constant factor. We present also the results of preliminary computational experiments, which confirm the superiority of the accelerated scheme.

An extension of the proximal point algorithm with Bregman distances on Hadamard manifolds

In this paper we present an extension of the proximal point algorithm with Bregman distances to solve constrained minimization problems with quasiconvex and convex objective function on Hadamard manifolds. The proposed algorithm is a modified and extended version of the one presented in Papa Quiroz and Oliveira (J Convex Anal 16(1): 49---69, 2009). An advantage of the proposed algorithm, for the nonconvex case, is that in each iteration the algorithm only needs to find a stationary point of the proximal function and not a global minimum. For that reason, from the computational point of view, the proposed algorithm is more practical than the earlier proximal method. Another advantage, for the convex case, is that using minimal condition on the problem data as well as on the proximal parameters we get the same convergence results of the Euclidean proximal algorithm using Bregman distances.

Anti-periodic solutions for nonlinear evolution equations

Abstract In this paper, we use the homotopy method to establish the existence and uniqueness of anti-periodic solutions for the nonlinear anti-periodic problem { x ˙ + A ( t , x ) + B x = f ( t ) a.e. t ∈ R , x ( t + T ) = − x ( t ) , where A ( t , x ) is a nonlinear map and B is a bounded linear operator from R N to R N . Sufficient conditions for the existence of the solution set are presented. Also, we consider the nonlinear evolution problems with a perturbation term which is multivalued. We show that, for this problem, the solution set is nonempty and weakly compact in W 1 , 2 ( I , R N ) for the case of convex valued perturbation and prove the existence theorems of anti-periodic solutions for the nonconvex case. All illustrative examples are provided.

Distributed stochastic power control in ad hoc networks: a nonconvex optimization case

Utility-based power allocation in wireless ad-hoc networks is inherently nonconvex because of the global coupling induced by the co-channel interference. To tackle this challenge, we first show that the globally optimal point lies on the boundary of the feasible region, which is utilized as a basis to transform the utility maximization problem into an equivalent max-min problem with more structure. By using extended duality theory, penalty multipliers are introduced for penalizing the constraint violations, and the minimum weighted utility maximization problem is then decomposed into subproblems for individual users to devise a distributed stochastic power control algorithm, where each user stochastically adjusts its target utility to improve the total utility by simulated annealing. The proposed distributed power control algorithm can guarantee global optimality at the cost of slow convergence due to simulated annealing involved in the global optimization. The geometric cooling scheme and suitable penalty parameters are used to improve the convergence rate. Next, by integrating the stochastic power control approach with the back-pressure algorithm, we develop a joint scheduling and power allocation policy to stabilize the queueing systems. Finally, we generalize the above distributed power control algorithms to multicast communications, and show their global optimality for multicast traffic.

Strong Fermat Rules for Constrained Set-Valued Optimization Problems on Banach Spaces

In this paper, we propose two kinds of optimality concepts, called the sharp minima and the weak sharp minima, for a constrained set-valued optimization problem. Subsequently, we extend the Fermat rules for the local minima of the constrained set-valued optimization problem to the sharp minima and the weak sharp minima in Banach spaces or Asplund spaces, by means of the Mordukhovich generalized differentiation and the normal cone. As applications, we investigate the generalized inequality systems with constraints, and get some characterizations of error bounds for the constrained generalized inequality systems in convex and nonconvex cases.

Improvement sets and vector optimization

In this paper we focus on minimal points in linear spaces and minimal solutions of vector optimization problems, where the preference relation is defined via an improvement set E. To be precise, we extend the notion of E-optimal point due to Chicco et al. in [4] to a general (non-necessarily Pareto) quasi ordered linear space and we study its properties. In particular, we relate the notion of improvement set with other similar concepts of the literature and we characterize it by means of sublevel sets of scalar functions. Moreover, we obtain necessary and sufficient conditions for E-optimal solutions of vector optimization problems through scalarization processes by assuming convexity assumptions and also in the general (nonconvex) case. By applying the obtained results to certain improvement sets we generalize well-known results of the literature referred to efficient, weak efficient and approximate efficient solutions of vector optimization problems.

Maximization of the Choquet integral over a convex set and its application to resource allocation problems

We study the problem of the Choquet integral maximization over a convex set. The problem is shown to be generally non-convex (and non-differentiable). We analyze the problem structure, and propose local and global search algorithms. The special case when the problem becomes concave is analyzed separately. For the non-convex case we propose a decomposition scheme which allows to reduce a non-convex problem to several concave ones. Decomposition is performed by finding the coarsest partition of a capacity into disjunction of totally monotone measures. We discuss its effectiveness and prove that the scheme is optimal for 2-additive capacities. An application of the developed methods to resource allocation problems concludes the article.