This paper presents a novel attribute selection approach for backprop neural networks. Previously, an attribute selection technique known as the wrapper model was shown effective for decision tree induction. However, it is prohibitively expensive when applied to real-world neural net training characterized by large volumes of data and many attribute choices. Our approach incorporates a weight analysis based heuristic called ANNIGMA to direct the search in the wrapper model and allows effective attribute selection feasible for neural net applications. Experimental results on standard data sets show that this approach can efficiently reduce the number of inputs while maintaining or even improving the accuracy. We also report two successful applications of our approach in the helicopter maintenance applications.
{"title":"A weight analysis-based wrapper approach to neural nets feature subset selection","authors":"D. Schuschel, Chun-Nan Hsu","doi":"10.1109/TAI.1998.744781","DOIUrl":"https://doi.org/10.1109/TAI.1998.744781","url":null,"abstract":"This paper presents a novel attribute selection approach for backprop neural networks. Previously, an attribute selection technique known as the wrapper model was shown effective for decision tree induction. However, it is prohibitively expensive when applied to real-world neural net training characterized by large volumes of data and many attribute choices. Our approach incorporates a weight analysis based heuristic called ANNIGMA to direct the search in the wrapper model and allows effective attribute selection feasible for neural net applications. Experimental results on standard data sets show that this approach can efficiently reduce the number of inputs while maintaining or even improving the accuracy. We also report two successful applications of our approach in the helicopter maintenance applications.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115335194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Scheduling techniques have been intensively studied by several research communities and have been applied to a wide range of applications in computer and manufacturing environments. Most of the scheduling problems are NP-hard. Therefore, heuristics and approximation algorithms must be used for large problems. Obviously these methods are of interest when they provide near optimal solutions and when computational complexity can be controlled. For this purpose, we have developed a method based on the Hopfield neural network model. This approach permits us to solve in an iterative way a scheduling problem, finding a solution through the minimization of an energy function. An interesting property of this approach is its capacity to trade-off the quality for computation time. Indeed, the convergence speed of the minimization process can be tuned by adapting several parameters that influence the quality of the results. By tuning these parameters, we can build a library of a set of run-time executions (contracts) of the Hopfield minimization process with different characteristics (quality, efficiency). We present two applications exploiting the advantage of having available anytime contract algorithms. The first application illustrates how to build a solution of a one machine scheduling problem within a delay that follows a stochastic distribution. The second application deals with unrelated parallel machine scheduling of non preemptive tasks.
{"title":"Stochastic and distributed anytime task scheduling","authors":"F. Charpillet, I. Chades, J. Gallone","doi":"10.1109/TAI.1998.744855","DOIUrl":"https://doi.org/10.1109/TAI.1998.744855","url":null,"abstract":"Scheduling techniques have been intensively studied by several research communities and have been applied to a wide range of applications in computer and manufacturing environments. Most of the scheduling problems are NP-hard. Therefore, heuristics and approximation algorithms must be used for large problems. Obviously these methods are of interest when they provide near optimal solutions and when computational complexity can be controlled. For this purpose, we have developed a method based on the Hopfield neural network model. This approach permits us to solve in an iterative way a scheduling problem, finding a solution through the minimization of an energy function. An interesting property of this approach is its capacity to trade-off the quality for computation time. Indeed, the convergence speed of the minimization process can be tuned by adapting several parameters that influence the quality of the results. By tuning these parameters, we can build a library of a set of run-time executions (contracts) of the Hopfield minimization process with different characteristics (quality, efficiency). We present two applications exploiting the advantage of having available anytime contract algorithms. The first application illustrates how to build a solution of a one machine scheduling problem within a delay that follows a stochastic distribution. The second application deals with unrelated parallel machine scheduling of non preemptive tasks.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121439646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
WWW servers have to serve many clients simultaneously and thus cannot provide intelligent services. We present an approach where intelligent problem solving is distributed so that compute-expensive tasks are carried out on the client side. To this end, we have implemented a library of constraint satisfaction techniques, called the JAVA constraint library, which allows composing applets that solve CSPs. We present the library and show several examples of applications.
{"title":"Distributing problem solving on the Web using constraint technology","authors":"Marc Torrens, R. Weigel, B. Faltings","doi":"10.1109/TAI.1998.744757","DOIUrl":"https://doi.org/10.1109/TAI.1998.744757","url":null,"abstract":"WWW servers have to serve many clients simultaneously and thus cannot provide intelligent services. We present an approach where intelligent problem solving is distributed so that compute-expensive tasks are carried out on the client side. To this end, we have implemented a library of constraint satisfaction techniques, called the JAVA constraint library, which allows composing applets that solve CSPs. We present the library and show several examples of applications.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"286 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122635849","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Diagnostic problems are proposed to be solved via the pattern recognition approach. The main example has been motivated by economic macro modelling, hence a reference model approach is adopted. The dynamic systems under consideration are assumed to be linear in three case studies and nonlinear in one case study. All four case studies illustrate the basic concept of this approach. Since the state vector represents the most compact information about the dynamic system, a measured/estimated state vector is chosen as the feature vector of a specific pattern vector. The results obtained verify our belief that this approach can be very useful if the assumption of linearity has been met. It is worthwhile to point out that although this approach appears fairly straightforward, it is quite powerful since computational power is now readily available. Since information systems with large databases are readily available, we believe the approach presented in the paper are certainly practical. The first case study deals with a macro economic model, while the second deals with a three state variable linear difference equation model. The third example illustrates use of patterns in frequency domain to make inference of a dynamic system diagnostic in time domain. Finally, the investigation focuses on the simulation of a nonlinear system. It is quite certain to conclude that the nonlinear system behavior is very dependent and sensitive to parameter changes.
{"title":"Diagnostics of dynamical systems by recognizing the default and abnormal pattern","authors":"Paul P. Wang, Mihir Rajopadhye","doi":"10.1109/TAI.1998.744864","DOIUrl":"https://doi.org/10.1109/TAI.1998.744864","url":null,"abstract":"Diagnostic problems are proposed to be solved via the pattern recognition approach. The main example has been motivated by economic macro modelling, hence a reference model approach is adopted. The dynamic systems under consideration are assumed to be linear in three case studies and nonlinear in one case study. All four case studies illustrate the basic concept of this approach. Since the state vector represents the most compact information about the dynamic system, a measured/estimated state vector is chosen as the feature vector of a specific pattern vector. The results obtained verify our belief that this approach can be very useful if the assumption of linearity has been met. It is worthwhile to point out that although this approach appears fairly straightforward, it is quite powerful since computational power is now readily available. Since information systems with large databases are readily available, we believe the approach presented in the paper are certainly practical. The first case study deals with a macro economic model, while the second deals with a three state variable linear difference equation model. The third example illustrates use of patterns in frequency domain to make inference of a dynamic system diagnostic in time domain. Finally, the investigation focuses on the simulation of a nonlinear system. It is quite certain to conclude that the nonlinear system behavior is very dependent and sensitive to parameter changes.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116663448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In recent work, a general framework for constraint satisfaction over finite domains has been proposed, based on the concept of semiring-valued constraints. Classical CSPs, fuzzy CSPs, and hierachical CSPs can be easily cast in this general framework. In this paper, we claim that translating any constraint problem into a semiring-based constraint problem makes it possible to express global information about the problem more easily, especially in the case of non-crisp or preference constraints. Applying this concept to the case of set-based semirings, we give a theoretical result and two practical applications developed using clp(FD, S), a full and efficient implementation of semiring-based constraint satisfaction in the CLP paradigm.
{"title":"Encoding global constraints in semiring-based constraint solving","authors":"Y. Georget, P. Codognet","doi":"10.1109/TAI.1998.744878","DOIUrl":"https://doi.org/10.1109/TAI.1998.744878","url":null,"abstract":"In recent work, a general framework for constraint satisfaction over finite domains has been proposed, based on the concept of semiring-valued constraints. Classical CSPs, fuzzy CSPs, and hierachical CSPs can be easily cast in this general framework. In this paper, we claim that translating any constraint problem into a semiring-based constraint problem makes it possible to express global information about the problem more easily, especially in the case of non-crisp or preference constraints. Applying this concept to the case of set-based semirings, we give a theoretical result and two practical applications developed using clp(FD, S), a full and efficient implementation of semiring-based constraint satisfaction in the CLP paradigm.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127011271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pattern recognition techniques for time-series forecasting are beginning to be realised as an important tool for predicting chaotic behaviour of dynamic systems. In this paper we develop the concept of a pattern modelling and recognition system which is used for predicting future behaviour of time-series using local approximation. In this paper we compare this forecasting tool with neural networks. We also study the effect of noise filtering on the performance of the proposed system. Fourier analysis is used for noise-filtering the time-series. The results show that Fourier analysis is an important tool for improving the performance of the proposed forecasting system. The results are discussed on three benchmark series and the real US S&P financial index.
{"title":"Dynamic time-series forecasting using local approximation","authors":"Sameer Singh, Paul McAtackney","doi":"10.1109/TAI.1998.744870","DOIUrl":"https://doi.org/10.1109/TAI.1998.744870","url":null,"abstract":"Pattern recognition techniques for time-series forecasting are beginning to be realised as an important tool for predicting chaotic behaviour of dynamic systems. In this paper we develop the concept of a pattern modelling and recognition system which is used for predicting future behaviour of time-series using local approximation. In this paper we compare this forecasting tool with neural networks. We also study the effect of noise filtering on the performance of the proposed system. Fourier analysis is used for noise-filtering the time-series. The results show that Fourier analysis is an important tool for improving the performance of the proposed forecasting system. The results are discussed on three benchmark series and the real US S&P financial index.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"42 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126045465","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A new parallel hybrid method for solving the satisfiability problem that combines cellular genetic algorithms and the random walk (WSAT) strategy of GSAT is presented. The method, called CGWSAT, uses a cellular genetic algorithm to perform a global search on a random initial population of candidate solutions and a local selective generation of new strings. Global search is specialized in local search by adopting the WSAT strategy. CGWSAT has been implemented on a Meiko CS-2 parallel machine using a two-dimensional cellular automaton as a parallel computation model. The algorithm has been tested on randomly generated problems and some classes of problems from the DIMACS test set.
{"title":"Combining cellular genetic algorithms and local search for solving satisfiability problems","authors":"G. Folino, C. Pizzuti, G. Spezzano","doi":"10.1109/TAI.1998.744842","DOIUrl":"https://doi.org/10.1109/TAI.1998.744842","url":null,"abstract":"A new parallel hybrid method for solving the satisfiability problem that combines cellular genetic algorithms and the random walk (WSAT) strategy of GSAT is presented. The method, called CGWSAT, uses a cellular genetic algorithm to perform a global search on a random initial population of candidate solutions and a local selective generation of new strings. Global search is specialized in local search by adopting the WSAT strategy. CGWSAT has been implemented on a Meiko CS-2 parallel machine using a two-dimensional cellular automaton as a parallel computation model. The algorithm has been tested on randomly generated problems and some classes of problems from the DIMACS test set.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115410158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We present a modification of the GTS (Generalized Takagi-Sugeno) model (Fiordaliso, 1996). The key idea is to constrain the conclusions of each rule to perform a convex combination of the input patterns. This constraint allows to interpret each rule as an input patterns mixer and therefore contributes to a better understanding of the system inference.
{"title":"Analysis improvement of Takagi-Sugeno fuzzy rules using convexity constraints","authors":"A. Fiordaliso","doi":"10.1109/TAI.1998.744848","DOIUrl":"https://doi.org/10.1109/TAI.1998.744848","url":null,"abstract":"We present a modification of the GTS (Generalized Takagi-Sugeno) model (Fiordaliso, 1996). The key idea is to constrain the conclusions of each rule to perform a convex combination of the input patterns. This constraint allows to interpret each rule as an input patterns mixer and therefore contributes to a better understanding of the system inference.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"128 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128214761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Second order hidden Markov models have been used for a long time in pattern recognition, especially in speech recognition. Their main advantages over other methods (neural networks etc.) are their capabilities of modeling noisy temporal signals of variable length. In a previous work, we proposed a new method based on second order hidden Markov models to learn and recognize places in an indoor environment by a mobile robot, and showed that this approach is well suited for learning and recognizing places. In this paper, we propose major modifications to increase the global rate of place recognition. Results of experiments on a real robot with distinctive places are given.
{"title":"Second order hidden Markov models for place recognition: new results","authors":"O. Aycard, Jean-François Mari, F. Charpillet","doi":"10.1109/TAI.1998.744879","DOIUrl":"https://doi.org/10.1109/TAI.1998.744879","url":null,"abstract":"Second order hidden Markov models have been used for a long time in pattern recognition, especially in speech recognition. Their main advantages over other methods (neural networks etc.) are their capabilities of modeling noisy temporal signals of variable length. In a previous work, we proposed a new method based on second order hidden Markov models to learn and recognize places in an indoor environment by a mobile robot, and showed that this approach is well suited for learning and recognizing places. In this paper, we propose major modifications to increase the global rate of place recognition. Results of experiments on a real robot with distinctive places are given.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129649357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kenneth M. F. Choi, Jimmy Ho-man Lee, Peter James Stuckey
Heuristic repair algorithms, a class of local search methods, demonstrate impressive efficiency in solving some large-scale and hard instances of constraint satisfaction problems (CSPs). We draw a surprising connection between heuristic repair techniques and the discrete Lagrange multiplier methods by transforming CSPs into zero-one constrained optimization problems. A Lagrangian-based search scheme LSDL is proposed. We show how GENET, a representative heuristic repair algorithm, can be reconstructed from LSDL. The dual viewpoint of GENET as a heuristic repair method and Lagrange multiplier method allows us to investigate variants of GENET from both perspectives. Benchmarking results confirm that first, our reconstructed GENET has the same fast convergence behavior as other GENET implementations reported in the literature, competing favourably with other state-of-the-art methods on a set of hard graph colouring problems. Second, our best variant, which combines techniques from heuristic repair and Lagrangian methods, is always more efficient than the reconstructed GENET, and can better it by an order of magnitude.
{"title":"A Lagrangian reconstruction of a class of local search methods","authors":"Kenneth M. F. Choi, Jimmy Ho-man Lee, Peter James Stuckey","doi":"10.1109/TAI.1998.744838","DOIUrl":"https://doi.org/10.1109/TAI.1998.744838","url":null,"abstract":"Heuristic repair algorithms, a class of local search methods, demonstrate impressive efficiency in solving some large-scale and hard instances of constraint satisfaction problems (CSPs). We draw a surprising connection between heuristic repair techniques and the discrete Lagrange multiplier methods by transforming CSPs into zero-one constrained optimization problems. A Lagrangian-based search scheme LSDL is proposed. We show how GENET, a representative heuristic repair algorithm, can be reconstructed from LSDL. The dual viewpoint of GENET as a heuristic repair method and Lagrange multiplier method allows us to investigate variants of GENET from both perspectives. Benchmarking results confirm that first, our reconstructed GENET has the same fast convergence behavior as other GENET implementations reported in the literature, competing favourably with other state-of-the-art methods on a set of hard graph colouring problems. Second, our best variant, which combines techniques from heuristic repair and Lagrangian methods, is always more efficient than the reconstructed GENET, and can better it by an order of magnitude.","PeriodicalId":424568,"journal":{"name":"Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122223513","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：