Statistical natural language processing (NLP) and evolutionary algorithms (EAs) are two very active areas of research which have been combined many times. In general, statistical models applied to deal with NLP tasks ...
详细信息
Statistical natural language processing (NLP) and evolutionary algorithms (EAs) are two very active areas of research which have been combined many times. In general, statistical models applied to deal with NLP tasks require designing specific algorithms to be trained and applied to process new texts. The development of such algorithms may be hard. This makes EAs attractive since they offer a general design, yet providing a high performance in particular conditions of application. In this article, we present a survey of many works which apply EAs to different NLP problems, including syntactic and semantic analysis, grammar induction, summaries and text generation, document clustering and machine translation. This review finishes extracting conclusions about which are the best suited problems or particular aspects within those problems to be solved with an evolutionary algorithm.
We employ the variational theory of optimal control problems and evolutionary algorithms to investigate the form finding of minimum compliance elastic structures. Mathematical properties of ground structure approaches...
详细信息
We employ the variational theory of optimal control problems and evolutionary algorithms to investigate the form finding of minimum compliance elastic structures. Mathematical properties of ground structure approaches are discussed with reference to arbitrary collections of structural elements. A numerical procedure based on a Breeder Genetic Algorithm is proposed for the shape optimization of discrete structural models. Several numerical applications are presented, showing the ability of the adopted search strategy in avoiding local optimal solutions. The proposed approach is validated against a parade of results available in the literature.
A feature model is a compact representation of the products of a software product line. The automated extraction of information from feature models is a thriving topic involving numerous analysis operations, technique...
详细信息
A feature model is a compact representation of the products of a software product line. The automated extraction of information from feature models is a thriving topic involving numerous analysis operations, techniques and tools. Performance evaluations in this domain mainly rely on the use of random feature models. However, these only provide a rough idea of the behaviour of the tools with average problems and are not sufficient to reveal their real strengths and weaknesses. In this article, we propose to model the problem of finding computationally hard feature models as an optimization problem and we solve it using a novel evolutionary algorithm for optimized feature models (ETHOM). Given a tool and an analysis operation, ETHOM generates input models of a predefined size maximizing aspects such as the execution time or the memory consumption of the tool when performing the operation over the model. This allows users and developers to know the performance of tools in pessimistic cases providing a better idea of their real power and revealing performance bugs. Experiments using ETHOM on a number of analyses and tools have successfully identified models producing much longer executions times and higher memory consumption than those obtained with random models of identical or even larger size. (C) 2013 Elsevier Ltd. All rights reserved.
This paper characterizes general optimization problems into four categories based oil the solution representation schemes, as they have been the key to the design of various evolutionary algorithms (EAs). Four EAs hav...
详细信息
This paper characterizes general optimization problems into four categories based oil the solution representation schemes, as they have been the key to the design of various evolutionary algorithms (EAs). Four EAs have been designed for different formulations with the aim of utilizing similar and generalized strategies for all of them. Several modifications to the existing EAs have been proposed and studied. First, a new tradeoff function-based mutation has been proposed that takes advantages of Cauchy, Gaussian, random as well as chaotic mutations. In addition, a generalized learning rule has also been proposed to ensure more thorough and explorative search. A theoretical analysis has been performed to establish the convergence of the learning rule. A theoretical study has also been performed in order to investigate the various aspects of the search strategy employed by the new tradeoff-based mutations. A more logical parameter tuning has been done by introducing the concept of orthogonal arrays in the EA experimentation. The use of noise-based tuning ensures the robust parameter tuning that enables the EAs to perform remarkably well in the further experimentations. The performance of the proposed EAs has been analyzed for different problems of varying complexities. The results prove the supremacy of the proposed EAs over other well-established strategies given in the literature. (c) 2007 Elsevier Ltd. All rights reserved.
Multiobjective evolutionary algorithms (MOEAs) are typically proposed, studied, and applied as monolithic blocks with a few numerical parameters that need to be set. Few works have studied how the algorithmic componen...
详细信息
Multiobjective evolutionary algorithms (MOEAs) are typically proposed, studied, and applied as monolithic blocks with a few numerical parameters that need to be set. Few works have studied how the algorithmic components of these evolutionary algorithms can be classified and combined to produce new algorithmic designs. The motivation for studies of this latter type stem from the development of flexible software frameworks and the usage of automatic algorithm configuration methods to find novel algorithm designs. In this paper, we propose an MOEA template and a new conceptual view of its components that surpasses existing frameworks in both number of algorithms that can be instantiated from the template and flexibility to produce novel algorithmic designs. We empirically demonstrate the flexibility of our proposed framework by automatically designing MOEAs for continuous and combinatorial optimization problems. The automatically designed algorithms are often able to outperform six traditional MOEAs from the literature, even after tuning their numerical parameters.
Multi-modality can cause serious problems for many optimisers, often resulting convergence to sub-optimal modes. Even when this is not the case, it is often useful to locate and memorise a range of modes in the design...
详细信息
Multi-modality can cause serious problems for many optimisers, often resulting convergence to sub-optimal modes. Even when this is not the case, it is often useful to locate and memorise a range of modes in the design space. This is because "optimal" decision parameter combinations may not actually be feasible when moving from a mathematical model emulating the real problem, to engineering an actual solution, making a range of disparate modal solutions of practical use. This paper builds upon our work on the use of a collection of localised search algorithms for niche/mode discovery which we presented at UKCI 2013 when using a collection of surrogate models to guide mode search. Here we present the results of using a collection of exploitative local evolutionary algorithms (EAs) within the same general framework. The algorithm dynamically adjusts its population size according to the number of regions it encounters that it believes contain a mode and uses localised EAs to guide the mode exploitation. We find that using a collection of localised EAs, which have limited communication with each other, produces competitive results with the current state-of-the-art multi-modal optimisation approaches on the CEC 2013 benchmark functions.
The class of foraging algorithms is a relatively new field based on mimicking the foraging behavior of animals, insects, birds or fish in order to develop efficient optimization algorithms. The artificial bee colony (...
详细信息
The class of foraging algorithms is a relatively new field based on mimicking the foraging behavior of animals, insects, birds or fish in order to develop efficient optimization algorithms. The artificial bee colony (ABC), the bees algorithm (BA), ant colony optimization (ACO), and bacterial foraging optimization algorithms (BFOA) are examples of this class to name a few. This work provides a complete performance assessment of the four mentioned algorithms in comparison to the widely known differential evolution (DE), genetic algorithms (GAs), harmony search (HS), and particle swarm optimization (PSO) algorithms when applied to the problem of unconstrained nonlinear continuous function optimization. To the best of our knowledge, most of the work conducted so far using foraging algorithms has been tested on classical functions. This work provides the comparison using the well-known CEC05 benchmark functions based on the solution reached, the success rate, and the performance rate. (C) 2011 Elsevier Inc. All rights reserved.
We present a time complexity analysis of the Opt-IA artificial immune system (AIS). We first highlight the power and limitations of its distinguishing operators (i.e., hypermutations with mutation potential and ageing...
详细信息
We present a time complexity analysis of the Opt-IA artificial immune system (AIS). We first highlight the power and limitations of its distinguishing operators (i.e., hypermutations with mutation potential and ageing) by analysing them in isolation. Recent work has shown that ageing combined with local mutations can help escape local optima on a dynamic optimisation benchmark function. We generalise this result by rigorously proving that, compared to evolutionary algorithms (EAs), ageing leads to impressive speed-ups on the standard CLIFFd benchmark function both when using local and global mutations. Unless the stop at first constructive mutation (FCM) mechanism is applied, we show that hypermutations require exponential expected runtime to optimise any function with a polynomial number of optima. If instead FCM is used, the expected runtime is at most a linear factor larger than the upper bound achieved for any random local search algorithm using the artificial fitness levels method. Nevertheless, we prove that algorithms using hypermutations can be considerably faster than EAs at escaping local optima. An analysis of the complete Opt-IA reveals that it is efficient on the previously considered functions and highlights problems where the use of the full algorithm is crucial. We complete the picture by presenting a class of functions for which Opt-IA fails with overwhelming probability while standard EAs are efficient. (C) 2019 Published by Elsevier B.V.
Freight transportation is important for the national economy in many countries. An efficient distribution of products within supply chains may lower the associated costs and allow setting competitive prices to increas...
详细信息
Freight transportation is important for the national economy in many countries. An efficient distribution of products within supply chains may lower the associated costs and allow setting competitive prices to increase the number of sales. Many supply chain players use the cross-docking terminals to facilitate the cargo distribution process. An effective scheduling of the arriving trucks at the cross-docking terminals is critical to ensure their timely service. A number of evolutionary algorithms have been developed to solve the truck scheduling problem, some of which apply strong mutation for altering solutions throughout the search process, while the rest rely on weak mutation without providing any justification for applying a specific mutation mechanism. This study performs a comprehensive comparative analysis of the strong and weak mutation mechanisms. Furthermore, a novel heuristic algorithm, which accounts for the truck service priority and the truck service order restrictions, is proposed for initializing the chromosomes and population. The truck scheduling problem at a cross-docking terminal is formulated as a mixed integer programming model, minimizing the total weighted truck service cost. An evolutionary Algorithm is designed to solve the problem. Two categories of the evolutionary Algorithm, one of which applies strong mutation, while the other one relies on weak mutation, are evaluated based on various performance indicators. Results demonstrate that deployment of weak mutation improves the objective function value at termination on average by 10.8% as compared with strong mutation without affecting the computational time substantially. The analysis also shows that weak mutation yields more diverse population. Moreover, the proposed heuristic for initializing the chromosomes and population outperforms the initialization mechanisms that are commonly used in the literature.
This paper builds the normal model of fitness sharing with proportionate selection on real-valued functions, and derives the dynamic formula to describe the evolution process of the population with the fitness sharing...
详细信息
This paper builds the normal model of fitness sharing with proportionate selection on real-valued functions, and derives the dynamic formula to describe the evolution process of the population with the fitness sharing. The normal modeling simulation is investigated on specific test functions, and experimental results illustrate that the normal model is able to describe exactly the dynamics of the fitness sharing EAs and is a good platform to study the behavior of the fitness sharing EAs with regard to niching radius. The experimental results of the normal modeling simulation and the fitness sharing EAs verify the dilemma in finding optimal niche radius to achieve both good niching convergence and niching efficiency, for which a hybrid scheme is proposed to carry out the niching task. (C) 2009 Elsevier B. V. All rights reserved.
暂无评论