Mostrando recursos 1 - 20 de 373.860

  1. Generation of structured process models using genetic algorithms

    Hartmut Pohlheim; Daimler Benz Ag; Peter Marenbach
    The design of structured mathematical models of processes in a certain level of abstraction defined by the given task appears to be difficult and time consuming even for experienced experts. This paper reports on a new method for the design of structured process models based on the metaphor of Genetic Programming. This new methodology allows the automatic generation of non-linear process models in a self-organizing way. Keywords:

  2. A hybrid genetic-quantitative method for risk-return optimisation of credit portfolios

    Frank Schlottmann; Detlef Seese
    This paper proposes a new combination of quantitative models and Genetic Algorithms for the task of optimising credit portfolios. Currently, quantitative portfolio credit risk models are used to calculate portfolio risk figures, e. g. expected losses, unexpected losses and risk contributions. Usually, this information is used for optimising the risk-return profile of the portfolio. We show that gradient-like optimisation methods based on risk contributions can lead to inefficient portfolio structures. To avoid this local optima problem, our optimisation method combines quantitative model features with Genetic Algorithms. The hybrid approach in this paper consists of a task specific Genetic Algorithm that...

  3. Abstract

    Partha Biswas; Sudarshan Banerjee; Laura Pozzi; Paolo Ienne; Nikil Dutt
    Customization of processor architectures through Instruction Set Extensions (ISEs) is an effective way to meet the growing performance demands of embedded applications. A high-quality ISE generation approach needs to obtain results close to those achieved by experienced designers, particularly for complex applications that exhibit regularity: expert designers are able to exploit manually such regularity in the data flow graphs to generate high-quality ISEs. In this paper, we present ISEGEN, an approach that identifies high-quality ISEs by iterative improvement following the basic principles of the well-known Kernighan-Lin (K-L) min-cut heuristic. Experimental results on a number of MediaBench, EEMBC and cryptographic applications...

  4. Vitri- A Generic Framework for Engineering Decision Support Systems on Heterogeneous Computer Networks

    W. Baugh; Sujay V Kumar
    Vitri is an object-oriented framework implemented in Java for high-performance distributed computing. Using Vitri, applications can engage in cooperative problem solving by dividing their tasks among heterogeneous clusters of workstations and PCs. Vitri’s features include basic support for distributed computing and communication, as well as visual tools for evaluating run-time performance, and modules for heuristic optimization. It balances loads dynamically using a client-side task pool, allows the addition or removal of servers during a run, and provides fault tolerance transparently for servers and networks. Among its more powerful features are modules for heuristic optimization and decision support tools such as...

  5. Parallel Genetic Algorithms on Programmable Graphics Hardware

    Qizhi Yu; Chongcheng Chen; Zhigeng Pan
    Parallel genetic algorithms are usually implemented on parallel machines or distributed systems. This paper describes how fine grained parallel genetic algorithms can be mapped to programmable graphics hardware found in commodity PC. Our approach stores chromosomes and their fitness values in texture memory on graphics card. Both fitness evaluation and genetic operations are implemented entirely with fragment programs executed on graphics processing unit in parallel. We demonstrate the effectiveness of our approach by comparing it with compatible software implementation. The presented approach allows us benefit from the advantages of parallel genetic algorithms on low-cost platform.

  6. Programming, called rule-based Genetic Programming, or

    In this paper we introduce a new approach for Genetic

  7. Multi-objective hybrid genetic algorithm for bicriteria network design problem

    Mitsuo Gen; Lin Lin
    This paper considers the Bicriteria Network Design Problem (bNDP) with the two conflicting objectives of minimizing cost and maximizing flow. Network design problems where even one flow measure be maximized, are often NP-hard problems. But, in real-life applications, it is often the case that the network to be built is required to optimize multi-criteria simultaneously. Thus the calculation of the multi-criteria network design problems is a difficult task. This paper proposes a new Multiobjective Hybrid Genetic Algorithm (mo-hGA) approach, and shows how the performance of multiobjective genetic algorithm (moGA) can be improved by hybridization with Fuzzy Logic Control (FLC) and...

  8. Evaluation of Injection Island GA Performance on Flywheel Design Optimization

    David Eby; R. C. Averill; William F. Punch Iii; Erik D. Goodman
    This paper first describes optimal design of elastic flywheels using an Injection Island Genetic Algorithm (iiGA). An iiGA in combination with a finite element code is used to search for shape variations to optimize the Specific Energy Density of flywheels (SED is the rotational energy stored per unit mass). iiGA’s seek solutions simultaneously at different levels of refinement of the problem representation (and correspondingly different definitions of the fitness function) in separate subpopulations (islands). Solutions are sought first at low levels of refinement with an axisymmetric plane stress finite element code for high-speed exploration of the coarse design space. Next,...

  9. An Introduction to Genetic Algorithms and Evolution Strategies Abstract – Genetic Algorithms and Evolution Strategies

    represent two of the three major Evolutionary Algorithms. This paper examines the history, theory and mathematical

  10. AASA: a Method of Automatically Acquiring Semantic Annotations

    Lixin Han; Guihai Chen; Li Xie; Lixin Han; Guihai Chen; Li Xie
    Abstract. An important precondition for the success of the Semantic Web is founded on the principle that the content of web pages will be semantically annotated. This paper proposes a method of automatically acquiring semantic annotations (AASA). In the AASA method, we employ a combination of data mining and optimization to acquire semantic annotations. Key features of AASA include combining association rules, inference mechanism, genetic algorithm and self-organizing map to create semantic annotations, and using the k-nearest-neighbor query combined with simulated annealing to maintain semantic annotations.

  11. Using Assembly Representations to Enable Evolutionary Design of Lego Structures

    Maxim Peysakhov; Short Title; William C. Regli
    This paper presents an approach to the automatic generation of electromechanical engineering designs. We apply Messy Genetic Algorithm optimization techniques to the evolution of assemblies composed of the Lego TM structures. Each design is represented as a labeled assembly graph and is evaluated based on a set of behavior and structural equations. The initial populations are generated at random and design candidates for subsequent generations are produced by user-specified selection techniques. Crossovers are applied by using cut and splice operators at the random points of the chromosomes; random mutations are applied to modify the graph with a certain low probability....

  12. Genetic Algorithm –High Dynamic Range image

    K. Sivakami Sunadri; Dr V. Sadasivam; S. Pradeepa
    Transmission and processing plays a major role. It would not be possible to retrieve information from satellite and medical images without the help of Image processing techniques. Image Enhancement is the art of examining images for identifying objects and judging their significance. The proposed work uses the concept of GA to find the image with best Signal to noise Ratio (SNR). The basic GA does not require extensive knowledge of the search space, such as likely solution bounds or functional derivatives. The present work presents a new method on rendering high dynamic range images on conventional displays. This algorithm is...

  13. Early eukaryote evolution based on mitochondrial gene order breakpoints

    David Sankoff; David Bryant; Mdlanie Deneault; B. Franz; Lang Gertraud Burger
    We present a general heuristic for the median problem for induced breakpoints on genomes with unequal gene con-tent and incorporate this into a routine for estimating op-timal gene orders for the ancestral genomes in a fixed phy-logeny. The routine is applied to a phylogenetic study of an up-to-date set of completely sequenced protist mitochon-drial genomes, confirming some of the recent sequence-based groupings which have been proposed and, conversely, con-firming the usefulness of the breakpoint method as a phylo-genetic tool even for small genomes. 1 Introduction. The origin and early diversification of the eukaryotes is one of the fundamental problems of...


    Eddie Cheng; Raymond P. Kleinberg; Serge G. Kruk; William A. Lindsey; Daniel; E. Steffy
    Abstract. This paper gives a combinatorial approach to solving the student exam scheduling problem. The problem is to generate schedules that satisfy hard constraints while minimizing soft constraint voilations. This problem is NP-Hard. The problem is decomposed into stages that include finding stable sets, weighted bipartite matchings, maximum flow, and pathfinding in hypergraphs. We describe our method and discuss our results and implementation. Keywords: Scheduling, NP-Hard, Combinatorial Optimization 1. Background The scheduling of exams is a common problem faced by educational institutions. Even in the simplest forms of this problem, determining if a feasible schedule exists is NP-Complete. Many others...

  15. A-life and musical composition: A brief survey

    Eduardo R Miranda; Peter M Todd
    There have been a number of interesting applications of A-Life in music, ranging from associating musical notes to the cells of cellular automata, to forging genotypes of musical parameters for generating music using genetic algorithms. From the three approaches surveyed in this paper, only the cultural approach allows for the study of the circumstances and mechanisms whereby music might originate and evolve in virtual communities of musicians and listeners. This approach considers musical systems in the context of the origins and evolution of cultural conventions that may emerge under a number of constraints, such as psychological, physiological and ecological constraints.

  16. > REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < Intelligent Content Aware Services in 3G Wireless Networks

    Flavio De Angelis; Student Member; Ibrahim Habib; Senior Member; Fabrizio Davide; Mahmoud Naghshineh; Fellow Member
    Abstract—In this paper, we address the problem of optimizing the delivery of multimedia services with different quality of service (QoS) requirements to mobile users. We assume that the network provides two distinct classes of service (CoS) to which users may subscribe: Premium, or Economy. Subscribers to the Premium service pay more for their connections but receive a higher level of quality measured by a set of parameters such as call blocking probability, coding rate, and format of the multimedia services. By optimizing the delivery of the multimedia services, we mean that the network guarantees that all users receive their agreed-upon...

  17. Migration policies and takeover times in genetic algorithms

    Erick Cantu-paz
    A speci cation of a parallel genetic algorithm (GA) with multiple populations includes the size and number of the populations (demes), the topology of the connections between the demes, the migration rate, and the policy to select emigrants and to replace existing individuals with incoming migrants. The objective of this paper is to study how the migration policy a ects the speed of convergence. The choices of migrants and the replacement of individuals are not often considered important parameters of parallel GAs. However, these choices a ect considerably the speed of convergence, which is important because excessively slow or fast...

  18. Evolutionary computational approaches to solving the multiple traveling salesman problem using a neighborhood attractor schema

    Donald Sofge; Alan Schultz; Kenneth De Jong
    This paper presents a variation of the Euclidean Traveling Salesman Problem (TSP), the Multiple Traveling Salesman Problem (MTSP), and compares a variety of evolutionary computation algorithms and paradigms for solving it. Techniques implemented, analyzed, and discussed herein with regard to MTSP include use of a neighborhood attractor schema (a variation on k-means clustering), the &quot;shrink-wrap &quot; algorithm for local neighborhood optimization, particle swarm optimization, Monte-Carlo optimization, and a range of genetic algorithms and evolutionary strategies.

  19. Automated analog circuit design using genetic algorithms

    Navid Azizi
    Analog circuits, while being replaced by digital circuits in many cases, remain very important in high-speed applications such as communications. Analog circuit synthesis is very challenging, and has traditionally been performed by specialists who have a wealth of experience and intuition. Recently, much progression has been made in automating analog circuit synthesis

  20. Alternative bloat control methods

    Liviu Panait; Sean Luke
    Abstract. Bloat control is an important aspect of evolutionary computation methods, such as genetic programming, which must deal with genomes of arbitrary size. We introduce three new methods for bloat control: Biased Multi-Objective Parsimony Pressure (BMOPP), the Waiting Room, and Death by Size. These methods are unusual approaches to bloat control, and are not only useful in various circumstances, but two of them suggest novel approaches to attack the problem. BMOPP is a more traditional parsimony-pressure style bloat control method, while the other two methods do not consider parsimony as part of the selection process at all, but instead penalize...

Aviso de cookies: Usamos cookies propias y de terceros para mejorar nuestros servicios, para análisis estadístico y para mostrarle publicidad. Si continua navegando consideramos que acepta su uso en los términos establecidos en la Política de cookies.