Pub Date : 2013-10-31DOI: 10.1142/S0218488513400217
Yingjie Li, P. H. F. Ng, S. Shiu
The techniques of fuzzy measure and fuzzy integral have been successfully applied in various real-world applications. The determination of fuzzy measures is the most difficult part in problem solving. Signed efficiency measure, which is a special kind of fuzzy measure with the best representation ability but the highest complexity, is even harder to determine. Some methodologies have been developed for solving this problem such as artificial neural networks (ANNs) and genetic algorithms (GAs). However, none of the existing methods can outperform the others with unique advantages. Thus, there is a strong need to develop a new technique for learning distinct signed efficiency measures from data. Extreme learning machine (ELM) is a new learning paradigm for training single hidden layer feed-forward networks (SLFNs) with randomly chosen input weights and analytically determined output weights. In this paper, we propose an ELM based algorithm for signed efficiency measure determination. Experimental comparisons demonstrate the effectiveness of the proposed method in both time and accuracy.
{"title":"Extreme learning machine for determining signed efficiency measure from data","authors":"Yingjie Li, P. H. F. Ng, S. Shiu","doi":"10.1142/S0218488513400217","DOIUrl":"https://doi.org/10.1142/S0218488513400217","url":null,"abstract":"The techniques of fuzzy measure and fuzzy integral have been successfully applied in various real-world applications. The determination of fuzzy measures is the most difficult part in problem solving. Signed efficiency measure, which is a special kind of fuzzy measure with the best representation ability but the highest complexity, is even harder to determine. Some methodologies have been developed for solving this problem such as artificial neural networks (ANNs) and genetic algorithms (GAs). However, none of the existing methods can outperform the others with unique advantages. Thus, there is a strong need to develop a new technique for learning distinct signed efficiency measures from data. Extreme learning machine (ELM) is a new learning paradigm for training single hidden layer feed-forward networks (SLFNs) with randomly chosen input weights and analytically determined output weights. In this paper, we propose an ELM based algorithm for signed efficiency measure determination. Experimental comparisons demonstrate the effectiveness of the proposed method in both time and accuracy.","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"122 1","pages":"131-142"},"PeriodicalIF":1.5,"publicationDate":"2013-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85352532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-10-31DOI: 10.1142/S021848851340014X
Klaus Neumann, Matthias Rolf, Jochen J. Steil
The application of machine learning methods in the engineering of intelligent technical systems often requires the integration of continuous constraints like positivity, monotonicity, or bounded curvature in the learned function to guarantee a reliable performance. We show that the extreme learning machine is particularly well suited for this task. Constraints involving arbitrary derivatives of the learned function are effectively implemented through quadratic optimization because the learned function is linear in its parameters, and derivatives can be derived analytically. We further provide a constructive approach to verify that discretely sampled constraints are generalized to continuous regions and show how local violations of the constraint can be rectified by iterative re-learning. We demonstrate the approach on a practical and challenging control problem from robotics, illustrating also how the proposed method enables learning from few data samples if additional prior knowledge about the problem is...
{"title":"RELIABLE INTEGRATION OF CONTINUOUS CONSTRAINTS INTO EXTREME LEARNING MACHINES","authors":"Klaus Neumann, Matthias Rolf, Jochen J. Steil","doi":"10.1142/S021848851340014X","DOIUrl":"https://doi.org/10.1142/S021848851340014X","url":null,"abstract":"The application of machine learning methods in the engineering of intelligent technical systems often requires the integration of continuous constraints like positivity, monotonicity, or bounded curvature in the learned function to guarantee a reliable performance. We show that the extreme learning machine is particularly well suited for this task. Constraints involving arbitrary derivatives of the learned function are effectively implemented through quadratic optimization because the learned function is linear in its parameters, and derivatives can be derived analytically. We further provide a constructive approach to verify that discretely sampled constraints are generalized to continuous regions and show how local violations of the constraint can be rectified by iterative re-learning. We demonstrate the approach on a practical and challenging control problem from robotics, illustrating also how the proposed method enables learning from few data samples if additional prior knowledge about the problem is...","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"19 1","pages":"35-50"},"PeriodicalIF":1.5,"publicationDate":"2013-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74148308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-10-31DOI: 10.1142/S0218488513400175
Wentao Mao, Jiucheng Xu, Shengjie Zhao, Mei Tian
Recently, extreme learning machines (ELMs) have been a promising tool in solving a wide range of regression and classification applications. However, when modeling multiple related tasks in which only limited training data per task are available and the dimension is low, ELMs are generally hard to get impressive performance due to little help from the informative domain knowledge across tasks. To solve this problem, this paper extends ELM to the scenario of multi-task learning (MTL). First, based on the assumption that model parameters of related tasks are close to each other, a new regularization-based MTL algorithm for ELM is proposed to learn related tasks jointly via simple matrix inversion. For improving the learning performance, the algorithm proposed above is further formulated as a mixed integer programming in order to identify the grouping structure in which parameters are closer than others, and finally an alternating minimization method is presented to solve this optimization. Experiments conducted on a toy problem as well as real-life data set demonstrate the effectiveness of the proposed MTL algorithm compared to the classical ELM and the standard MTL algorithm.
{"title":"RESEARCH OF MULTI-TASK LEARNING BASED ON EXTREME LEARNING MACHINE","authors":"Wentao Mao, Jiucheng Xu, Shengjie Zhao, Mei Tian","doi":"10.1142/S0218488513400175","DOIUrl":"https://doi.org/10.1142/S0218488513400175","url":null,"abstract":"Recently, extreme learning machines (ELMs) have been a promising tool in solving a wide range of regression and classification applications. However, when modeling multiple related tasks in which only limited training data per task are available and the dimension is low, ELMs are generally hard to get impressive performance due to little help from the informative domain knowledge across tasks. To solve this problem, this paper extends ELM to the scenario of multi-task learning (MTL). First, based on the assumption that model parameters of related tasks are close to each other, a new regularization-based MTL algorithm for ELM is proposed to learn related tasks jointly via simple matrix inversion. For improving the learning performance, the algorithm proposed above is further formulated as a mixed integer programming in order to identify the grouping structure in which parameters are closer than others, and finally an alternating minimization method is presented to solve this optimization. Experiments conducted on a toy problem as well as real-life data set demonstrate the effectiveness of the proposed MTL algorithm compared to the classical ELM and the standard MTL algorithm.","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"19 1","pages":"75-85"},"PeriodicalIF":1.5,"publicationDate":"2013-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74238846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-10-31DOI: 10.1142/S0218488513400163
Dong Huang, Zhiqi Shen
The design of fuzzy cognitive maps (FCMs) mainly relies on human knowledge, which implies subjectivity of the developed model. This affects the accuracy of an FCM significantly. In order to address this issue, we propose a novel learning model for FCMs in this paper. It achieves efficient learning by automatically adjusting the system parameters according to the environment. The learning model consists of extreme learning machine (ELM) and a curious model, where ELM learns from the modeled system and the curious model helps to further improve the performance of ELM. We use an example to illustrate the effectiveness of our model. The simulation results show that our model helps to improve the accuracy of FCMs.
{"title":"A CURIOUS LEARNING MODEL WITH ELM FOR FUZZY COGNITIVE MAPS","authors":"Dong Huang, Zhiqi Shen","doi":"10.1142/S0218488513400163","DOIUrl":"https://doi.org/10.1142/S0218488513400163","url":null,"abstract":"The design of fuzzy cognitive maps (FCMs) mainly relies on human knowledge, which implies subjectivity of the developed model. This affects the accuracy of an FCM significantly. In order to address this issue, we propose a novel learning model for FCMs in this paper. It achieves efficient learning by automatically adjusting the system parameters according to the environment. The learning model consists of extreme learning machine (ELM) and a curious model, where ELM learns from the modeled system and the curious model helps to further improve the performance of ELM. We use an example to illustrate the effectiveness of our model. The simulation results show that our model helps to improve the accuracy of FCMs.","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"34 1","pages":"63-74"},"PeriodicalIF":1.5,"publicationDate":"2013-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75601403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-08-12DOI: 10.1142/S0218488513400102
Z. Qin, D. Wang, Xiang Li
In practice, security returns cannot be accurately predicted due to lack of historical data. Therefore, statistical methods and experts' experience are always integrated to estimate future security returns, which are hereinafter regarded as random fuzzy variables. Random fuzzy variable is a powerful tool to deal with the portfolio optimization problem including stochastic parameters with ambiguous expected returns. In this paper, we first define the semivariance of random fuzzy variable and prove its several properties. By considering the semivariance as a risk measure, we establish the mean-semivariance models for portfolio optimization problem with random fuzzy returns. We design a hybrid algorithm with random fuzzy simulation to solve the proposed models in general cases. Finally, we present a numerical example and compare the results to illustrate the mean-semivariance model and the effectiveness of the algorithm.
{"title":"MEAN-SEMIVARIANCE MODELS FOR PORTFOLIO OPTIMIZATION PROBLEM WITH MIXED UNCERTAINTY OF FUZZINESS AND RANDOMNESS","authors":"Z. Qin, D. Wang, Xiang Li","doi":"10.1142/S0218488513400102","DOIUrl":"https://doi.org/10.1142/S0218488513400102","url":null,"abstract":"In practice, security returns cannot be accurately predicted due to lack of historical data. Therefore, statistical methods and experts' experience are always integrated to estimate future security returns, which are hereinafter regarded as random fuzzy variables. Random fuzzy variable is a powerful tool to deal with the portfolio optimization problem including stochastic parameters with ambiguous expected returns. In this paper, we first define the semivariance of random fuzzy variable and prove its several properties. By considering the semivariance as a risk measure, we establish the mean-semivariance models for portfolio optimization problem with random fuzzy returns. We design a hybrid algorithm with random fuzzy simulation to solve the proposed models in general cases. Finally, we present a numerical example and compare the results to illustrate the mean-semivariance model and the effectiveness of the algorithm.","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"74 1","pages":"127-139"},"PeriodicalIF":1.5,"publicationDate":"2013-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83747817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-08-12DOI: 10.1142/S0218488513400023
Lixing Yang, Xiaofei Yang, C. You
Focusing on finding a pre-specified basis path in a network, this research formulates a two-stage stochastic optimization model for the least expected time shortest path problem, in which random scenario-based time-invariant link travel times are utilized to capture the uncertainty of the realworld traffic network. In this model, the first stage aims to find a basis path for the trip over all the scenarios, and the second stage intends to generate the remainder path adaptively when the realizations of random link travel times are updated after a pre-specified time threshold. The GAMS optimization software is introduced to find the optimal solution of the proposed model. The numerical experiments demonstrate the performance of the proposed approaches.
{"title":"STOCHASTIC SCENARIO-BASED TIME-STAGE OPTIMIZATION MODEL FOR THE LEAST EXPECTED TIME SHORTEST PATH PROBLEM","authors":"Lixing Yang, Xiaofei Yang, C. You","doi":"10.1142/S0218488513400023","DOIUrl":"https://doi.org/10.1142/S0218488513400023","url":null,"abstract":"Focusing on finding a pre-specified basis path in a network, this research formulates a two-stage stochastic optimization model for the least expected time shortest path problem, in which random scenario-based time-invariant link travel times are utilized to capture the uncertainty of the realworld traffic network. In this model, the first stage aims to find a basis path for the trip over all the scenarios, and the second stage intends to generate the remainder path adaptively when the realizations of random link travel times are updated after a pre-specified time threshold. The GAMS optimization software is introduced to find the optimal solution of the proposed model. The numerical experiments demonstrate the performance of the proposed approaches.","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"103 1 1","pages":"17-33"},"PeriodicalIF":1.5,"publicationDate":"2013-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86265643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-08-12DOI: 10.1142/S0218488513400084
Chao Wang, Minghu Ha
In order to deal with learning problems of random set samples encountered in real-world, according to random set theory and convex quadratic programming, a new support vector machine based on random set samples is constructed. Experimental results show that the new support vector machine is feasible and effective.
{"title":"SUPPORT VECTOR MACHINE BASED ON RANDOM SET SAMPLES","authors":"Chao Wang, Minghu Ha","doi":"10.1142/S0218488513400084","DOIUrl":"https://doi.org/10.1142/S0218488513400084","url":null,"abstract":"In order to deal with learning problems of random set samples encountered in real-world, according to random set theory and convex quadratic programming, a new support vector machine based on random set samples is constructed. Experimental results show that the new support vector machine is feasible and effective.","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"23 1","pages":"101-112"},"PeriodicalIF":1.5,"publicationDate":"2013-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74559461","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-08-12DOI: 10.1142/S0218488513400059
Yaodong Ni, Qiaoni Shi
In this paper, we study the problem of targeting a set of individuals to trigger a cascade of influence in a social network such that the influence diffuses to all individuals with the minimum time, given that the cost of initially influencing each individual is with randomness and that the budget available is limited. We adopt the incremental chance model to characterize the diffusion of influence, and propose three stochastic programming models that correspond to three different decision criteria respectively. A modified greedy algorithm is presented to solve the proposed models, which can flexibly trade off between solution performance and computational complexity. Experiments are performed on random graphs, by which we show that the algorithm we present is effective.
{"title":"MINIMIZING THE COMPLETE INFLUENCE TIME IN A SOCIAL NETWORK WITH STOCHASTIC COSTS FOR INFLUENCING NODES","authors":"Yaodong Ni, Qiaoni Shi","doi":"10.1142/S0218488513400059","DOIUrl":"https://doi.org/10.1142/S0218488513400059","url":null,"abstract":"In this paper, we study the problem of targeting a set of individuals to trigger a cascade of influence in a social network such that the influence diffuses to all individuals with the minimum time, given that the cost of initially influencing each individual is with randomness and that the budget available is limited. We adopt the incremental chance model to characterize the diffusion of influence, and propose three stochastic programming models that correspond to three different decision criteria respectively. A modified greedy algorithm is presented to solve the proposed models, which can flexibly trade off between solution performance and computational complexity. Experiments are performed on random graphs, by which we show that the algorithm we present is effective.","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"274 1","pages":"63-74"},"PeriodicalIF":1.5,"publicationDate":"2013-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75094227","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-08-12DOI: 10.1142/S0218488513400035
Mei Yu, Jiangze Bian, Haibin Xie, Qin Zhang, D. Ralescu
In this paper, we employ the resampling method to reduce the sample errors and increase the robustness of the classic mean variance model. By comparing the performances of the classic mean variance portfolio and the resampled portfolio, we show that the resampling method can enhance the investment efficiency. Through an empirical study of Chinese investors who invest in both Chinese market and other twelve major financial markets, we show that the resampling method helps to improve the performance of the mean variance model.
{"title":"STUDY ON THE RESAMPLING TECHNIQUE FOR RISK MANAGEMENT IN THE INTERNATIONAL PORTFOLIO SELECTION BASED ON CHINESE INVESTORS","authors":"Mei Yu, Jiangze Bian, Haibin Xie, Qin Zhang, D. Ralescu","doi":"10.1142/S0218488513400035","DOIUrl":"https://doi.org/10.1142/S0218488513400035","url":null,"abstract":"In this paper, we employ the resampling method to reduce the sample errors and increase the robustness of the classic mean variance model. By comparing the performances of the classic mean variance portfolio and the resampled portfolio, we show that the resampling method can enhance the investment efficiency. Through an empirical study of Chinese investors who invest in both Chinese market and other twelve major financial markets, we show that the resampling method helps to improve the performance of the mean variance model.","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"9 1","pages":"35-49"},"PeriodicalIF":1.5,"publicationDate":"2013-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81506466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-08-12DOI: 10.1142/S0218488513400096
Yufu Ning, Li-mei Yan, Huanbin Sha
A model is constructed for a type of multi-period inventory problem with deteriorating items, in which demands are assumed to be uncertain variables. The objective is to minimize the expected total cost including the ordering cost, inventory holding cost and deteriorating cost under constraints that demands should be satisfied with some service level in each period. To solve the model, two methods are proposed in different cases. When uncertain variables are linear, a crisp equivalent form of the model is provided. For the general cases, a hybrid algorithm integrating the 99-method and genetic algorithm is designed. Two examples are given to illustrate the effectiveness of the model and solving methods.
{"title":"A MULTI-PERIOD INVENTORY MODEL FOR DETERIORATING ITEMS IN UNCERTAIN ENVIRONMENT","authors":"Yufu Ning, Li-mei Yan, Huanbin Sha","doi":"10.1142/S0218488513400096","DOIUrl":"https://doi.org/10.1142/S0218488513400096","url":null,"abstract":"A model is constructed for a type of multi-period inventory problem with deteriorating items, in which demands are assumed to be uncertain variables. The objective is to minimize the expected total cost including the ordering cost, inventory holding cost and deteriorating cost under constraints that demands should be satisfied with some service level in each period. To solve the model, two methods are proposed in different cases. When uncertain variables are linear, a crisp equivalent form of the model is provided. For the general cases, a hybrid algorithm integrating the 99-method and genetic algorithm is designed. Two examples are given to illustrate the effectiveness of the model and solving methods.","PeriodicalId":50283,"journal":{"name":"International Journal of Uncertainty Fuzziness and Knowledge-Based Systems","volume":"62 1","pages":"113-125"},"PeriodicalIF":1.5,"publicationDate":"2013-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90941555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：