Search

In order to improve the evaluation effect of classroom education, this paper proposes the MFO intelligent optimization algorithm based on the idea of machine learning, and builds the classroom education effect evaluation model based on the MFO intelligent optimization algorithm. Moreover, this paper uses a logarithmic spiral to simulate the path of the moth to the flame and invert the pending parameters in the mathematical model, and adds vertical and horizontal algorithms and chaos operators on this basis. The crisscross algorithm allows different moth individuals and the same moth to perform cross calculations with different computing dimensions to increase the diversity of moth populations, so that moths in the search space can traverse the entire search space as much as possible to find a better solution. Moreover, in view of the problems of BP neural network such as low fitting accuracy, this paper applies the CCMFO algorithm to improve the BP neural network to form the CCMFO-BP algorithm, and improves the weight and threshold update process of the BP neural network to make the network operation more efficient and accurate. Finally, this paper designs experiments to analyze the performance of the model constructed in this paper. The research results show that the model constructed in this paper meets the expected requirements. [ABSTRACT FROM AUTHOR]

Nowadays, the existence of continuous optimization problems has led researchers to come up with a variety of methods to solve continues optimization problems. The metaheuristic algorithms are one of the most popular and common ways to solve continuous optimization problems. Firefly Algorithm (FA) is a successful metaheuristic algorithm for solving continuous optimization problems; however, although this algorithm performs very well in local search, it has weaknesses and disadvantages in finding solution in global search. This problem has caused this algorithm to be trapped locally and the balance between exploration and exploitation cannot be well maintained. In this paper, three different approaches based on the Dragonfly Algorithm (DA) processes and the OBL method are proposed to improve exploration, performance, efficiency and information-sharing of the FA and to avoid the FA getting stuck in local trap. In the first proposed method (FADA), the robust processes of DA are used to improve the exploration, performance and efficiency of the FA; and the second proposed method (OFA) uses an Opposition-Based Learning (OBL) algorithm to accelerate the convergence and exploration of the FA. Finally, in the third approach, which is referred to as OFADA in this paper, a hybridization of the hybrid FADA and the OBL method is used to improve the convergence and accuracy of the FA. The three proposed methods were implemented on functions with 2, 4, 10, and 30 dimensions. The results of the implementation of these three proposed methods showed that OFADA approach outperformed the other two proposed methods and other compared metaheuristic algorithms in different dimensions. In addition, all the three proposed methods provided better results compared with other metaheuristic algorithms on small-dimensional functions. However, performance of many metaheuristic algorithms decreased with increasing the dimensions of the functions. While the three proposed methods, in particular the OFADA approach, have been able to make better converge with the higher-dimensional optimization functions toward the target in comparison with other metaheuristic algorithms, and to show a high performance. [ABSTRACT FROM AUTHOR]

In this paper, a problem of scheduling jobs with different sizes and fuzzy processing times (FPT) on non-identical parallel batch machines to minimize makespan is investigated. Moreover, the processing time (PT) of each batch is subject to the location-based learning and total-PT-based deterioration effect. Since this is an NP-hard combinatorial optimization problem, an improved intelligent algorithm based on fruit fly optimization algorithm (IFOA) is proposed. To verify the performance of the algorithm, the IFOA is compared with three state-of-the-art algorithms. The comparative results demonstrate that the proposed IFOA outperforms the other compared algorithms. [ABSTRACT FROM AUTHOR]

This paper is devoted to the study of an extension of dynamic programming approach which allows optimization of partial decision rules relative to the length or coverage. We introduce an uncertainty measure J(T) which is the difference between number of rows in a decision table T and number of rows with the most common decision for T. For a nonnegative real number γ, we consider γ-decision rules (partial decision rules) that localize rows in subtables of T with uncertainty at most γ. Presented algorithm constructs a directed acyclic graph Δγ(T) which nodes are subtables of the decision table T given by systems of equations of the kind 'attribute = value'. This algorithm finishes the partitioning of a subtable when its uncertainty is at most γ. The graph Δγ(T) allows us to describe the whole set of so-called irredundant γ-decision rules. We can optimize such set of rules according to length or coverage. This paper contains also results of experiments with decision tables from UCI Machine Learning Repository. [ABSTRACT FROM AUTHOR]

Temporal difference learning and eligibility traces are two mechanisms for solving reinforcement learning problems. The temporal difference technique bootstraps the state value or state-action value at every step as with dynamic programming, and learns by sampling episodes from experience as in the Monte Carlo approach. Eligibility traces is a mechanism that offers a means for recording the degree of which state is eligible for undergoing learning process. This paper aims to investigate the underlying mechanism of eligibility traces strategies using on-policy and off-policy learning algorithms. In doing so, the performance metrics can be obtained by defining the learning problem in a simulation environment, in conjunction with different learning algorithms. However, measuring learning performance and analysing sensibility are very expensive because such performance metrics can only be obtained by running an experiment with different parameter values. This paper proposes a comparative study for analysing the mechanism of eligibility traces. The objective of this paper is to compare and investigate the influences on performance caused by those different approaches. [ABSTRACT FROM AUTHOR]

Hybrid Evolutionary Algorithms are a promising alternative to deal with the problem of selecting the most appropriate Evolutionary Algorithm for a specific problem. By means of the combination of different heuristic optimization approaches, it is possible to profit from the benefits of the best approach or, even more, to discover synergies between the algorithms that could improve the results of the best performing individual algorithm. Nowadays, there is an active research in the design of dynamic or adaptive combination strategies for hybrid algorithms. However, little research has been done in the automatic learning of the best hybridization strategy. This paper proposes a new methodology for developing intelligent adaptive hybrid algorithms that uses data mining techniques to analyze the results from past executions. The proposed methodology has been evaluated on a well-known benchmark on continuous optimization made up of 19 different functions and several dimensions (50, 100, 200 and 500). Several analyses have been conducted and statistical tests have been used for validating the results. The generated hybrid algorithm has achieved outstanding results, obtaining significantly better results than the MOS algorithm, the most performant algorithm on this benchmark, and the CMA-ES algorithm, one of the reference algorithms in continuous optimization. [ABSTRACT FROM AUTHOR]

Many real-world problems, like microchip design, can be modeled by means of the well-known traveling salesman problem (TSP). Many instances of this problem can be found in the literature. Although several optimization algorithms have been applied to TSP instances, the selection of the more promising algorithm is, in practice, a difficult decision. In this paper, a new meta-learning-based approach is investigated for the selection of optimization algorithms for TSP instances. Essentially, a learning model is trained with TSP instances for which the performance of a set of optimization algorithms is known a priori. Then, the learned model is used to predict the best algorithm for a new TSP instance. Each instance is described by meta-features that capture characteristics of the TSP that affect the performance of the optimization algorithms. Given that the best solution for a given TSP instance can be obtained by several algorithms, the meta-learning problem is considered here to be a multi-label classification problem. Several experiments illustrate the performance of the proposed approach, with promising results. [ABSTRACT FROM AUTHOR]

Simulated Annealing (SA) has been adopted by many Ensemble Clustering methods to achieve global combinational optimisation. However the performance of SA is sensitive to the settings of its parameters. Much work has been done for optimising the settings of these parameters over the last two decades, but few of them analysed the behaviour of different cooling functions for Ensemble Clustering. Our work has demonstrated that the clustering results could be invalid if we use SA for Ensemble Clustering without a good understanding of the behaviour of cooling functions. Therefore this paper aims to present the findings of how different cooling functions may affect the performance of Ensemble Clustering methods that use SA. We analyse the effect of cooling functions from three aspects: the convergence rate, the final value of the objective function, and the accuracy of results. Ten different cooling functions are tested on two Ensemble Clustering methods, and thirteen different datasets have been used for the experiments. The findings are particularly helpful for those who are interested in Ensemble Clustering methods as well as those who want to obtain a deep understanding of the behaviour of the cooling functions. [ABSTRACT FROM AUTHOR]

It is difficult to determine the recognition mechanism in the neurons of a neural network trained for pattern recognition due to the non-linear nature of neural networks. This paper describes a recognition mechanism of a four-layer back propagation neural network using Alopex algorithm. We have developed a small-scale, four-layered neural network model for simple character recognition, which can recognize the patterns transformed by affined conversion. Alopex algorithm is an iterative and stochastic processing method, which was proposed for optimization of a given cost function. In this case the receptive fields of the neurons in the output layers are obtained using the Alopex algorithm. [ABSTRACT FROM AUTHOR]

In this paper we investigate an algorithmic extension to the technique of Stacking for regression that prunes the ensemble set before application based on a consideration of the training accuracy and diversity of the ensemble members. We evaluate two variants of this approach in comparison to the standard Stacking algorithm, one of which is a static approach that prunes back the ensemble to the same constant size; the other of which is a variable approach prunes the ensemble to an appropriate level based on measures of accuracy and diversity of the ensemble members. We show that on average both techniques are robust in performance to their non-pruned counterpart, while having the advantage of producing smaller and less complex ensembles. In the latter respect, the static approach proved more effective, but we show that the variable approach lends itself better for further optimization. [ABSTRACT FROM AUTHOR]