In this paper, we propose a new service platform which provides stable QoS by allocating components dynamically with considering fault-tolerance. We assume that many kinds of components are running on component server...
详细信息
The branch of complex system spans over a wide range of areas from physical and technological systems to social and biological systems. Hemodynamics is a branch of physiology and is a complex system which deals with t...
详细信息
Tactile information is valuable in determining properties of objects that are inaccessible from visual perception. In this work, we present a tactile perception strategy that allows any mobile robot with tactile senso...
详细信息
In vehicular ad-hoc networks (VANETs), moving vehicles carry data and exchange it as they pass each other. Storage in a vehicle, in other words, the amount of saved data is limited resource in general. Therefore, attr...
详细信息
In vehicular ad-hoc networks (VANETs), moving vehicles carry data and exchange it as they pass each other. Storage in a vehicle, in other words, the amount of saved data is limited resource in general. Therefore, attributes of data that should be carried by vehicles is an important factor in order to efficiently disseminate desired data. In this paper, we focus on popularity of data and propose probabilistic data deployment based on its popularity. The proposed method deploys the data to vehicles randomly according to its relative popularity. By this simple operation, data can be deployed inside a vehicular network and as a result, data with high popularity tends to have larger number of copies inside the network than lower-popularity data. Performance evaluation shows that our deployment can work well in limited storage.
Recently evolutionary multiobjective optimization (EMO) algorithms have been frequently used for the design of fuzzy rule-based systems. Such a study is often referred to as multiobjective genetic fuzzy systems (MoGFS...
详细信息
Recently evolutionary multiobjective optimization (EMO) algorithms have been frequently used for the design of fuzzy rule-based systems. Such a study is often referred to as multiobjective genetic fuzzy systems (MoGFS). Whereas a large number of interesting results on MoGFS have already been reported in the literature, the search ability of EMO algorithms in MoGFS is not necessarily high because fuzzy system design is formulated as large-scale combinatorial optimization problems with many decision variables. In this paper, we show that simple changes in problem formulations of MoGFS often lead to large differences in obtained non-dominated fuzzy rule-based systems. Our idea for improving the search ability of EMO algorithms is to use multiple weighted sums with different weight vectors instead of original objectives. This idea is applicable not only to MoGFS but also to SoGFS (single-objective genetic fuzzy systems). When our idea is used in SoGFS, a two-objective problem is formulated as a weighted sum of an additional term with a small weight and the original objective. We examine the effectiveness of our idea through computational experiments on classification problems. It is clearly shown that simple changes in multiobjective problems in MoGFS often lead to large accuracy improvement.
Evolutionary multiobjective optimization (EMO) algorithms have often been used to search for a number of non-dominated fuzzy rule-based classifiers with respect to their accuracy and complexity. It is, however, pointe...
详细信息
Evolutionary multiobjective optimization (EMO) algorithms have often been used to search for a number of non-dominated fuzzy rule-based classifiers with respect to their accuracy and complexity. It is, however, pointed out in some studies that the entire accuracy-complexity tradeoff surface is not always found by well-known and frequently-used EMO algorithms such as NSGA-II. Especially it is very difficult for EMO algorithms to find fuzzy rule-based classifiers with high accuracy around the edge of the tradeoff surface. One simple idea for the design of accurate fuzzy rule-based classifiers is the use of fine fuzzy partitions with a number of small antecedent fuzzy sets. The use of fine fuzzy partitions usually improves the accuracy of fuzzy rule-based classifiers on training data. It may, however, have some side-effects such as the deterioration of classification accuracy on test data and the increase in the search space for fuzzy system design. In this paper, we examine the use of fine fuzzy partitions in the evolutionary multiobjective design of fuzzy rule-based classifiers. Experimental results show that the use of fine fuzzy partitions almost always increases the number of obtained non-dominated fuzzy rule-based classifiers, almost always improve their training data accuracy, and often improve their test data accuracy for some data sets. We also examine the relation between the granularity of fuzzy partitions and the number of antecedent conditions (i.e., rule length).
Genetic fuzzy rule selection has been successfully used to design accurate and interpretable fuzzy classifiers from numerical data. Its computation time, however, becomes very long when it is applied to large data set...
详细信息
Genetic fuzzy rule selection has been successfully used to design accurate and interpretable fuzzy classifiers from numerical data. Its computation time, however, becomes very long when it is applied to large data sets. To drastically decrease the computation time of genetic fuzzy rule selection without severely degrading the accuracy of obtained fuzzy classifiers, we proposed its parallel distributed implementation on parallel computers with multiple CPU cores. The main feature of our implementation is to divide not only a population but also training data into multiple sub-groups. In this paper, we try to further decrease the computation time by dividing the training data into very small data subsets. While the number of data subsets is the same as the number of CPU cores in our former study, we divide the training data into much more data subsets than CPU cores. Through computational experiments, we examine the effects of using very small data subsets with frequent rotation on the computation time and the accuracy of obtained fuzzy classifiers.
The objective of this work is to embed watermark information into digital audio data as the deterioration of sound quality is not perceivable to human ears. Hence, we consider that watermark information is embedded in...
详细信息
The objective of this work is to embed watermark information into digital audio data as the deterioration of sound quality is not perceivable to human ears. Hence, we consider that watermark information is embedded in the consonance to melody line or bass line. Generally, unison, octave, perfect fifth and perfect fourth are called consonance and human ears can hardly feel annoyed for consonance. Hence, we focus on perfect fifth and perfect fourth in the consonance. For decision of embedding positions, we need to estimate tone pitches, and we use spectrogram analysis as a technique of automatic music transcription. We embed watermark information with perfect fifth or perfect fourth of consonance from the pitch which is estimated by spectrogram analysis, and we aim at good sound quality of watermarked audio signal without sense of discomfort. Therefore, we propose an audio watermarking method by using automatic music transcription information.
This paper investigates varying the decoder weight of the language model (LM) when translating different parts of a sentence. We determine the condition under which the LM weight should be adapted. We find that a bett...
详细信息
This paper investigates varying the decoder weight of the language model (LM) when translating different parts of a sentence. We determine the condition under which the LM weight should be adapted. We find that a better translation can be achieved by varying the LM weight when decoding the most problematic spot in a sentence, which we refer to as a difficult segment. Two adaptation strategies are proposed and compared through experiments. We find that adapting a different LM weight for every difficult segment resulted in the largest improvement in translation quality.
暂无评论