The cardinality-constrained portfolio optimization problem is NP-hard. Its Pareto front (or the Efficient Frontier - EF) is usually calculated by stochastic algorithms, including EAs. However, in certain cases the EF ...
详细信息
ISBN:
(纸本)9781450349390
The cardinality-constrained portfolio optimization problem is NP-hard. Its Pareto front (or the Efficient Frontier - EF) is usually calculated by stochastic algorithms, including EAs. However, in certain cases the EF may be decomposed into a union of sub-EFs. In this work we propose a systematic process of excluding sub-EFs dominated by others, enabling us to calculate non-dominated sub-EFs. We then calculate whole EFs to a high degree of accuracy for small cardinalities, providing an alternative to EAs in those cases. We can use also this to provide insight into EAs on the problem.
Directed graphs are widely used to model data flow and execution dependencies in streaming applications. This enables the utilization of graph partitioning algorithms for the problem of parallelizing computation for m...
详细信息
Our cybersecurity tool, RIVALS, develops adaptive network defense strategies by modeling adversarial network attack and defense behavior in peer-to-peer networks via coevolutionary algorithms. Currently, RIVALS DOS at...
详细信息
ISBN:
(纸本)9781450349390
Our cybersecurity tool, RIVALS, develops adaptive network defense strategies by modeling adversarial network attack and defense behavior in peer-to-peer networks via coevolutionary algorithms. Currently, RIVALS DOS attacks are modestly modeled by the selection of a node that is completely disabled for a resource-limited duration. Defenders have three different network routing protocols. Attack or mission completion and resource cost metrics serve as attacker and defender objectives. This work also includes a description of RIVALS' suite of coevolutionary algorithms that explore archiving as a means of maintaining progressive exploration and support the evaluation of different solution concepts. To compare and contrast the effectiveness of each algorithm, we execute simulations on 3 different network topologies. Our experiments show that it is possible to forgo the assurance of monotonically increasing results and still retain high quality results.
Relaxed forms of Pareto dominance have been shown to be the most effective way in which evolutionary algorithms can progress towards the Pareto-optimal front with a widely spread distribution of solutions. A popular c...
详细信息
ISBN:
(纸本)9783642198922
Relaxed forms of Pareto dominance have been shown to be the most effective way in which evolutionary algorithms can progress towards the Pareto-optimal front with a widely spread distribution of solutions. A popular concept is the epsilon-dominance technique, which has been employed as an archive update strategy in some multiobjective evolutionary algorithms. In spite of the great usefulness of the epsilon-dominance concept, there are still difficulties in computing an appropriate value of epsilon that provides the desirable number of nondominated points. Additionally, several viable solutions may be lost depending on the hypergrid adopted, impacting the convergence and the diversity of the estimate set. We propose the concept of cone epsilon-dominance, which is a variant of the epsilon-dominance, to overcome these limitations. Cone epsilon-dominance maintains the good convergence properties of epsilon-dominance, provides a better control over the resolution of the estimated Pareto front, and also performs a better spread of solutions along the front. Experimental validation of the proposed cone epsilon-dominance shows a significant improvement in the diversity of solutions over both the regular Pareto-dominance and the epsilon-dominance.
作者：
Jiang, DazhiFan, ZhunShantou Univ
Dept Comp Sci Shantou 515063 Peoples R China Shantou Univ
Guangdong Prov Key Lab Digital Signal & Image Pro Shantou 515063 Peoples R China Shantou Univ
Elect & Informat Engn Dept Shantou 515063 Peoples R China
At present there is a wide range of evolutionary algorithms available to researchers and practitioners. Despite the great diversity of these algorithms, virtually all of the algorithms share one feature: they have bee...
详细信息
At present there is a wide range of evolutionary algorithms available to researchers and practitioners. Despite the great diversity of these algorithms, virtually all of the algorithms share one feature: they have been manually designed. A fundamental question is "are there any algorithms that can design evolutionary algorithms automatically?" A more complete definition of the question is "can computer construct an algorithm which will generate algorithms according to the requirement of a problem?" In this paper, a novel evolutionary algorithm based on automatic designing of genetic operators is presented to address these questions. The resulting algorithm not only explores solutions in the problem space like most traditional evolutionary algorithms do, but also automatically generates genetic operators in the operator space. In order to verify the performance of the proposed algorithm, comprehensive experiments on 23 well-known benchmark optimization problems are conducted. The results show that the proposed algorithm can outperform standard differential evolution algorithm in terms of convergence speed and solution accuracy which shows that the algorithm designed automatically by computers can compete with the algorithms designed by human beings.
—Bilevel optimization is defined as a mathematical program, where an optimization problem contains another optimization problem as a constraint. These problems have received significant attention from the mathematica...
详细信息
This paper presents general finite impulse response (FIR) digital filter design with asymmetric coefficients to approximate passband and stopband magnitude responses and constant passband group delay specifications us...
详细信息
This paper presents general finite impulse response (FIR) digital filter design with asymmetric coefficients to approximate passband and stopband magnitude responses and constant passband group delay specifications using an evolutionary optimization algorithm called the Interactive Self Learning Algorithm (ISLA). Lowpass and bandpass digital filters are chosen and their design results are shown to demonstrate the effectiveness of the approach. Results indicate that passband and stopband peak magnitude errors and passband peak group delay error can be designed to approximate given specifications.
Container Loading Problems (CLPs) deal with determination of the optimal pattern for packing boxes into a given container usually with respect to the maximal utilization of the total container volume. On the other han...
详细信息
ISBN:
(纸本)9781450349390
Container Loading Problems (CLPs) deal with determination of the optimal pattern for packing boxes into a given container usually with respect to the maximal utilization of the total container volume. On the other hand, it is also important to maximize the utilization of the maximal container weight for which is paid when buying a shipment service. In this paper we analyze two genetic algorithms specially adopted to solve CLP. One of them is based on the Genetic Algorithm (GA) and is suitable to solve single-objective CLPs, while another one is based on the Non-dominated Sorting Genetic Algorithm (NSGA-II), suitable for solution of CLP by simultaneously considering both of the above mentioned objectives. The algorithms have been experimentally investigated by solving various CLP instances of different complexity. The obtained results showed that simultaneous consideration of both objectives using the proposed multi-objective optimization algorithm gives better results in utilization of container volume when solving complex CLP instances.
A large number of infeasible solutions often occur in population of evolutionary Computation (EC) solving the constraint combinatorial optimization problems. The greater the number of infeasible solutions in the popul...
详细信息
A large number of infeasible solutions often occur in population of evolutionary Computation (EC) solving the constraint combinatorial optimization problems. The greater the number of infeasible solutions in the population, the worse the performance of ECto search the solution, in the worst case, the algorithm ceases to run. The existing methods, penalty function or multi-objective optimization, can relieve partly the worst case of EC to run. However, they are actually to restrain the infeasible solutions surviving in population, the performance of the EC is not improved. In this study we propose an approach using an important feature of the infeasible solutions in Genetic algorithms (GA). The approach can not only solve the problem of algorithm ceases to run, but also improve effectively the performance of genetic algorithms searching the optimal solution. From examination of the proposed method on multidimensional knapsack problems, the application of method is effective to solve the problem of algorithm ceases to run as well as to improve clearly the performance of GA.
This paper demonstrates the structural optimization using evolutionary algorithms in a chalcogenide glass waveguide. Four features are taken into consideration while optimizing the waveguide structure, they include: s...
详细信息
ISBN:
(纸本)9780819488459
This paper demonstrates the structural optimization using evolutionary algorithms in a chalcogenide glass waveguide. Four features are taken into consideration while optimizing the waveguide structure, they include: single-mode, low dispersion, high nonlinearity and low loss. A set of waveguide structures which meet the design criteria are shown in the paper. The best structure enhances the nonlinear coefficient to 26000 /W/km at telecom wavelength. In this work, we demonstrate the methodology used to optimize waveguide as well as the procedure of conducting the experiment.
暂无评论