GUIDED HYBRID GENETIC ALGORITHM FOR SOLVING GLOBAL OPTIMIZATION PROBLEMS

Context. One of the leading problems in the world of artificial intelligence is the optimization of complex systems, which is often represented as a nonlinear function that needs to be minimized. Such functions can be multimodal, non-differentiable, and even set as a black box. Building effective methods for solving global optimization problems raises great interest among scientists. Objective. Development of a new hybrid genetic algorithm for solving global optimization problems, which is faster than existing analogues. Methods. One of the crucial challenges for hybrid methods in solving nonlinear global optimization problems is the rational use of local search, as its application is accompanied by quite expensive computational costs. This paper proposes a new GBOHGA hybrid genetic algorithm that reproduces guided local search and combines two successful modifications of genetic algorithms. The first one is BOHGA that establishes a qualitative balance between local and global search. The second one is HGDN that prevents reexploration of the previously explored areas of a search space. In addition, a modified bump-function and an adaptive scheme for determining one of its parameters – the radius of the “deflation” of the objective function in the vicinity of the already found local minimum – were presented to accelerate the algorithm. Results. GBOHGA performance compared to other known stochastic search heuristics on a set of 33 test functions in 5 and 25dimensional spaces. The results of computational experiments indicate the competitiveness of GBOHGA, especially in problems with multimodal functions and a large number of variables. Conclusions. The new GBOHGA hybrid algorithm, developed on the basis of the integration of guided local search ideas and BOHGA and HGDN algorithms, allows to save significant computing resources and speed up the solution process of the global optimization problem. It should be used to solve global optimization problems that arise in engineering design, solving organizational and management problems, especially when the mathematical model of the problem is complex and multidimensional.

PE-DCA: Penalty elimination based data center allocation technique using guided local search for IaaS cloud

In Cloud computing the user requests are passaged to data centers (DCs) to accommodate resources. It is essential to select the suitable DCs as per the user requests so that other requests should not be penalized in terms of time and cost. The searching strategies consider the execution time rather than the related penalties while searching DCs. In this work, we discuss Penalty Elimination-based DC Allocation (PE-DCA) using Guided Local Search (GLS) mechanism to locate suitable DCs with reduced cost, response time, and processing time. The PE-DCA addresses, computes, and eliminates the penalties involved in the cost and time through iterative technique using the defined objective and guide functions. The PE-DCA is implemented using CloudAnalyst with various configurations of user requests and DCs. We examine the PE-DCA and the execution after-effects of various costs and time parameters to eliminate the penalties and observe that the proposed mechanism performs best.

Facility Location with Tree Topology and Radial Distance Constraints

Let Gd=V,Ed be an input disk graph with a set of facility nodes V and a set of edges Ed connecting facilities in V. In this paper, we minimize the total connection cost distances between a set of customers and a subset of facility nodes S⊆V and among facilities in S, subject to the condition that nodes in S simultaneously form a spanning tree and an independent set according to graphs G¯d and Gd, respectively, where G¯d is the complement of Gd. Four compact polynomial formulations are proposed based on classical and set covering p-Median formulations. However, the tree to be formed with S is modelled with Miller–Tucker–Zemlin (MTZ) and path orienteering constraints. Example domains where the proposed models can be applied include complex wireless and wired network communications, warehouse facility location, electrical power systems, water supply networks, and transportation networks, to name a few. The proposed models are further strengthened with clique valid inequalities which can be obtained in polynomial time for disk graphs. Finally, we propose Kruskal-based heuristics and metaheuristics based on guided local search and simulated annealing strategies. Our numerical results indicate that only the MTZ constrained models allow obtaining optimal solutions for instances with up to 200 nodes and 1000 users. In particular, tight lower bounds are obtained with all linear relaxations, e.g., less than 6% for most of the instances compared to the optimal solutions. In general, the MTZ constrained models outperform path orienteering ones. However, the proposed heuristics and metaheuristics allow obtaining near-optimal solutions in significantly short CPU time and tight feasible solutions for large instances of the problem.

Automatic Verification of FSA Strategies via Counterexample-Guided Local Search for Invariants

Strategy representation and reasoning has received much attention over the past years. In this paper, we consider the representation of general strategies that solve a class of (possibly infinitely many) games with similar structures, and their automatic verification, which is an undecidable problem. We propose to represent a general strategy by an FSA (Finite State Automaton) with edges labelled by restricted Golog programs. We formalize the semantics of FSA strategies in the situation calculus. Then we propose an incomplete method for verifying whether an FSA strategy is a winning strategy by counterexample-guided local search for appropriate invariants. We implemented our method and did experiments on combinatorial game and also single-agent domains. Experimental results showed that our system can successfully verify most of them within a reasonable amount of time.