Dynamical system is a mathematical approach to model the nonlinear dynamics of complex systems over space and time. A causality-informed genetic programming (GP) ensemble methodology has been proposed recently by the ...
详细信息
ISBN:
(纸本)9781450363396
Dynamical system is a mathematical approach to model the nonlinear dynamics of complex systems over space and time. A causality-informed genetic programming (GP) ensemble methodology has been proposed recently by the authors to automatically infer dynamical systems from system observations. The method adopts a variable decomposition approach relies on expert defined causal models. However, in practice these models are bound to have inconsistencies due to human involvement. Hence, in this paper we evaluate the sensitivity of the ensemble method to the accuracy of input causal models that are used as ground truth in the formation of the ensemble. This is done by varying the accuracy of known causal models through introducing deliberate noise in models' causal relationships. Three benchmark problems are used to evaluate the performance of the proposed methodology where the output of different ensembles is compared with a standard GP algorithm. The empirical results show the effectiveness of the proposed methodology in inferring closely matching target equations under different levels of noise and learning better models than the standard GP algorithm in most cases.
Purpose: This work presents a task-driven joint optimization of fluence field modulation (FFM) and regularization in quadratic penalized-likelihood (PL) reconstruction. Conventional FFM strategies proposed for filtere...
详细信息
ISBN:
(数字)9781510607101
ISBN:
(纸本)9781510607095;9781510607101
Purpose: This work presents a task-driven joint optimization of fluence field modulation (FFM) and regularization in quadratic penalized-likelihood (PL) reconstruction. Conventional FFM strategies proposed for filtered-backprojection (FBP) are evaluated in the context of PL reconstruction for comparison. Methods: We present a task-driven framework that leverages prior knowledge of the patient anatomy and imaging task to identify FFM and regularization. We adopted a maxi-min objective that ensures a minimum level of detectability index (d') across sample locations in the image volume. The FFM designs were parameterized by 2D Gaussian basis functions to reduce dimensionality of the optimization and basis function coefficients were estimated using the covariance matrix adaptation evolutionary strategy (CMA-ES) algorithm. The FFM was jointly optimized with both space-invariant and spatially-varying regularization strength (beta) - the former via an exhaustive search through discrete values and the latter using an alternating optimization where beta was exhaustively optimized locally and interpolated to form a spatially-varying map. Results: The optimal FFM inverts as beta increases, demonstrating the importance of a joint optimization. For the task and object investigated, the optimal FFM assigns more fluence through less attenuating views, counter to conventional FFM schemes proposed for FBP. The maxi-min objective homogenizes detectability throughout the image and achieves a higher minimum detectability than conventional FFM strategies. Conclusions: The task-driven FFM designs found in this work are counter to conventional patterns for FBP and yield better performance in terms of the maxi-min objective, suggesting opportunities for improved image quality and/or dose reduction when model-based reconstructions are applied in conjunction with FFM.
Differential Evolution (DE) has been considered as an effective approach for solving numerical optimization problems. Due to different characteristics of optimization problems, many proposed algorithms try to perform ...
详细信息
ISBN:
(纸本)9781538605271
Differential Evolution (DE) has been considered as an effective approach for solving numerical optimization problems. Due to different characteristics of optimization problems, many proposed algorithms try to perform consistently over a range of problems. The proposed algorithm combines both LSHADE and MTS-LS1 by giving both a participation ratio of the fitness evaluation where each technique works until finishing its participation budget or reaching the optimum solution. Using hybrid model gives an opportunity to achieve better performance for both algorithms. The evaluation of this algorithm has been tested using CEC 2014 benchmark problems.
Optimization algorithms, especially evolutionary algorithms, have gained wide acceptance among many disciplines such as electrical, control or industrial engineering. The ability to solve an objective or cost function...
详细信息
ISBN:
(纸本)9781538617236
Optimization algorithms, especially evolutionary algorithms, have gained wide acceptance among many disciplines such as electrical, control or industrial engineering. The ability to solve an objective or cost function with more unknown parameters than known equations, which make the problem unsolvable by means of deterministic approaches, is the main benefit of using evolutionary algorithms. In all cases optimization algorithms rely on starting with a completely random initial solution set then evolves this set towards better ones in respect to fitness or objective function iteratively. In this study, we have proven that starting from a unique point after a brief local and deterministic search instead of a pure random set is more beneficial in respect to fitness function evaluation count, or computation time. This approach, although it can be applied to any optimization algorithm, is a natural add-on to Big Bang - Big Crunch (BBBC) optimization method.
We investigate two popular trajectory-based algorithms from biology and physics to answer a question of general significance: when is it beneficial to reject improvements? A distinguishing factor of SSWM (Strong Selec...
详细信息
ISBN:
(纸本)9781450349208
We investigate two popular trajectory-based algorithms from biology and physics to answer a question of general significance: when is it beneficial to reject improvements? A distinguishing factor of SSWM (Strong Selection Weak Mutation), a popular model from population genetics, compared to the Metropolis algorithm (MA), is that the former can reject improvements, while the latter always accepts them. We investigate when one strategy outperforms the other. Since we prove that both algorithms converge to the same stationary distribution, we concentrate on identifying a class of functions inducing large mixing times, where the algorithms will outperform each other over a long period of time. The outcome of the analysis is the definition of a function where SSWM is efficient, while Metropolis requires at least exponential time.
Metaheuristics for optimization based on the immune network theory are often highlighted by being able to maintain the diversity of candidate solutions present in the population, allowing a greater coverage of the sea...
详细信息
ISBN:
(纸本)9781450349208
Metaheuristics for optimization based on the immune network theory are often highlighted by being able to maintain the diversity of candidate solutions present in the population, allowing a greater coverage of the search space. This work, however, shows that algorithms derived from the aiNET family for the solution of combinatorial problems may not present an adequate strategy for search space exploration, leading to premature convergence in local minimums. In order to solve this issue, a hybrid metaheuristic called VNS-aiNET is proposed, integrating aspects of the COPT-aiNET algorithm with characteristics of the trajectory metaheuristic Variable Neighborhood Search (VNS), as well as a new fitness function, which makes it possible to escape from local minima and enables it to a greater exploration of the search space. The proposed metaheuristic is evaluated using a scheduling problem widely studied in the literature. The performed experiments show that the proposed hybrid metaheuristic presents a convergence superior to two approaches of the aiNET family and to the reference algorithms of the literature. In contrast, the solutions present in the resulting immunological memory have less diversity when compared to the aiNET family approaches.
This study presents a new sine and cosine (S&C) optimization algorithm using a novel position update approach. In the proposed algorithm, the position update procedure for each search agent is determined by two co...
详细信息
ISBN:
(纸本)9781538608043
This study presents a new sine and cosine (S&C) optimization algorithm using a novel position update approach. In the proposed algorithm, the position update procedure for each search agent is determined by two coefficients, namely the exploration rate and the exploitation rate. These coefficients are updated in each run of the algorithm and provide an appropriate balance between the exploration and exploitation phases. The performances of the proposed algorithm and the sine cosine algorithm (SCA) were evaluated on a set of benchmark functions. The results indicate that in addition to a faster convergence speed, the S&C algorithm achieved the global best with a higher accuracy.
This paper is the results of research about the weather forecast in Bandung Regency using one of the evolutionary algorithms (EA), that is Genetic Programming (GP). In this research, we use the monthly rainfall data i...
详细信息
ISBN:
(纸本)9781538616673
This paper is the results of research about the weather forecast in Bandung Regency using one of the evolutionary algorithms (EA), that is Genetic Programming (GP). In this research, we use the monthly rainfall data in Bandung Regency for the last 11 years (2005-2015). First of all, the data is processed by Weighted Moving Average (WMA) algorithm as preprocessing step. Next, GP Algorithm is used to process the rainfall weather forecast which represents non-linear chromosome as a tree. In a population, chromosomes have different lengths because a child's chromosomes can be longer or shorter than his parents. To produce child, GP Algorithm applies the recombination process and the mutation using the several scenarios of probability of crossover and probability of mutation. By applying Genetic Programming algorithm, the system of weather forecast in Bandung regency has a performance above 70% in accuracy.
In this paper, we study the impact of using a hybrid-technique approach, which is a combination of genetic algorithm (GA) and protein's free energy minimization calculations, to predict protein tertiary structure....
详细信息
ISBN:
(纸本)9781538611913
In this paper, we study the impact of using a hybrid-technique approach, which is a combination of genetic algorithm (GA) and protein's free energy minimization calculations, to predict protein tertiary structure. We compare the results with a basic approach which applies genetic algorithm only. A genetic algorithm is used to predict the protein structure using the primary structure, the amino acids sequence of a given polypeptide chain, as input. After that, we combine the GA with energy minimization feature. Finally, the outcomes of both experiments are analyzed. Results reveal that the hybrid approach outperforms the basic one.
Ensemble clustering can improve the generalization ability of a single clustering algorithm and generate a more robust clustering result by integrating multiple base clusterings, so it becomes the focus of current clu...
详细信息
Ensemble clustering can improve the generalization ability of a single clustering algorithm and generate a more robust clustering result by integrating multiple base clusterings, so it becomes the focus of current clustering research. Ensemble clustering aims at finding a consensus partition which agrees as much as possible with base clusterings. Genetic algorithm is a highly parallel, stochastic, and adaptive search algorithm developed from the natural selection and evolutionary mechanism of biology. In this paper, an improved genetic algorithm is designed by improving the coding of chromosome. A new membrane evolutionary algorithm is constructed by using genetic mechanisms as evolution rules and combines with the communication mechanism of cell-like P system. The proposed algorithm is used to optimize the base clusterings and find the optimal chromosome as the final ensemble clustering result. The global optimization ability of the genetic algorithm and the rapid convergence of the membrane system make membrane evolutionary algorithm performbetter than several state-of-the-art techniques on six real-world UCI data sets.
暂无评论