Search

This paper presents a derivative-free nonmonotone hybrid tabu search to compute a solution of overdetermined systems of inequalities and equalities through the global optimization of an appropriate merit function. The proposed algorithm combines global and local searches aiming to reduce computational effort. Preliminary numerical results show the effectiveness of the combined heuristic. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new global optimization algorithm for non-smooth unconstrained optimization problems. We construct a new global optimization (GO) method based on the function new smoothing technique. We give some numerical examples in order to demonstrate the effectiveness of our algorithm. [ABSTRACT FROM AUTHOR]

Solving global optimization problems plays a key role in most branches of natural sciences whilst dealing with everyday problems. With the help of optimization algorithms and the exponentially fast growth of the capabilities of the underlying hardware, the scale of feasible optimization tasks is reaching new levels. We help this goal with revisiting GLOBAL, a stochastic optimization method aiming to solve non-linear constrained optimization problems by the penalty function approach. It is a versatile tool for a broad range of problems, proven to be competitive in multiple comparisons. With the similarly rapid evolution of programming habits and tools, time naturally passed by GLOBAL’s latest implementation so that it became a drawback. Now, we present GlobalJ, a fully modularized Java framework refurbishing and extending the potential of this algorithm. [ABSTRACT FROM AUTHOR]

We propose a new method for solving univariate global optimization problems by combining a lower bound function given in (αBB) method [1] [For more details (Algorithms,Methods) one see [1]], with the improved lower bound function of the method developed in [2]. The new lower bound function is better than the two lower bound functions by its construction. The complementarity of the two lower bound functions allows us to derive the convex/concave test and the pruning step which accelerate the convergence of the proposed method. Illustrative examples are treated efficiently. [ABSTRACT FROM AUTHOR]