The Experts below are selected from a list of 771096 Experts worldwide ranked by ideXlab platform
E. Servat - One of the best experts on this subject based on the ideXlab platform.
-
Simulation of runoff in West Africa: Is there a single data-Model Combination that produces the best simulation results?
Journal of Hydrology, 2008Co-Authors: A. Dezetter, S. Girard, J.e. Paturel, Gilles Mahe, S. Ardoin-bardin, E. ServatAbstract:Is there a single Combination of hydrologic Model and data that yields the best simulations of runoff from a set of catchments in West Africa in the coming decades? To answer this question, a semi-distributed Modelling platform was developed. It comprises two Models (GR2M and Water Balance Model) and various datasets as inputs to these Models (three potential evapotranspiration [PET] grids and four soil water holding capacity [WHC] grids). The platform makes it possible to regionalise runoff flows, which are discretised in grids of mesh size 0.5 · 0.5. On the basis of these datasets and Models, we compare the performance of the various possible data-Model Combinations. The study area contains 49 catchments located in Cote d'Ivoire, Guinea, Mali, Burkina Faso and Niger. The analysis shows that the selected Models are hardly sensitive at all to the different PET grids but more sensitive to the soil grids. The GR2M Model clearly gives the best results for the set of catchments, but does not perform well over the entire study area, whatever the grid dataset used: It is thus seen to be difficult to define a single data-Model Combination that is optimal for simulating runoff in West Africa. As to the existence of a general' Model that could be used anywhere, we find that, for the selected study area, neither of the two Models selected fully meets the objective.
-
Simulation of runoff in West Africa: Is there a single data-Model Combination that produces the best simulation results?
Journal of Hydrology, 2008Co-Authors: A. Dezetter, S. Girard, J.e. Paturel, Gilles Mahe, S. Ardoin-bardin, E. ServatAbstract:Is there a single Combination of hydrologic Model and data that yields the best simulations of runoff from a set of catchments in West Africa in the coming decades? To answer this question, a semi-distributed Modelling platform was developed. It comprises two Models (GR2M and Water Balance Model) and various datasets as inputs to these Models (three potential evapotranspiration [PET] grids and four soil water holding capacity [WHC] grids). The platform makes it possible to regionalise runoff flows, which are discretised in grids of mesh size 0.5 degrees x 0.5 degrees. On the basis of these datasets and Models, we compare the performance of the various possible data-Model Combinations. The study area contains 49 catchments located in Cote d'lvoire, Guinea, Mali, Burkina Faso and Niger. The analysis shows that the selected Models are hardly sensitive at all to the different PET grids but more sensitive to the soil grids. The GR2M Model clearly gives the best results for the set of catchments, but does not perform well over the entire study area, whatever the grid dataset used: It is thus seen to be difficult to define a single data-Model Combination that is optimal for simulating runoff in West Africa. As to the existence of a "general" Model that could be used anywhere, we find that, for the selected study area, neither of the two Models selected fully meets the objective
Shizhong Liao - One of the best experts on this subject based on the ideXlab platform.
-
PRICAI - Model Combination for support vector regression via regularization path
Lecture Notes in Computer Science, 2012Co-Authors: Mei Wang, Shizhong LiaoAbstract:In order to improve the generalization performance of support vector regression (SVR), we propose a novel Model Combination method for SVR on regularization path. First, we construct the initial candidate Model set using the regularization path, whose inherent piecewise linearity makes the construction easy and effective. Then, we elaborately select the Models for Combination from the initial Model set through the improved Occam's Window method and the input-dependent strategy. Finally, we carry out the Combination on the selected Models using the Bayesian Model averaging. Experimental results on benchmark data sets show that our Combination method has significant advantage over the Model selection methods based on generalized cross validation (GCV) and Bayesian information criterion (BIC). The results also verify that the improved Occam's Window method and the input-dependent strategy can enhance the predictive performance of the Combination Model.
-
Probabilistic Model Combination for Support Vector Machine Using Positive-Definite Kernel-Based Regularization Path
Advances in Intelligent and Soft Computing, 2011Co-Authors: Ning Zhao, Zhihui Zhao, Shizhong LiaoAbstract:Model Combination is an important approach to improving the generalization performance of support vector machine (SVM), but usually has low computational efficiency. In this paper, we propose a novel probabilistic Model Combination method for support vector machine on regularization path (PMCRP). We first design an efficient regularization path algorithm, namely the regularization path of support vector machine based on positive-definite kernel (PDSVMP), which constructs the initial candidate Model set. Then, we combine the initial Models using Bayesian Model averaging. Experimental results on benchmark datasets show that PMCRP has significant advantage over cross-validation and the Generalized Approximate Cross-Validation (GACV), meanwhile guaranteeing high computation efficiency of Model Combination.
A. Dezetter - One of the best experts on this subject based on the ideXlab platform.
-
Simulation of runoff in West Africa: Is there a single data-Model Combination that produces the best simulation results?
Journal of Hydrology, 2008Co-Authors: A. Dezetter, S. Girard, J.e. Paturel, Gilles Mahe, S. Ardoin-bardin, E. ServatAbstract:Is there a single Combination of hydrologic Model and data that yields the best simulations of runoff from a set of catchments in West Africa in the coming decades? To answer this question, a semi-distributed Modelling platform was developed. It comprises two Models (GR2M and Water Balance Model) and various datasets as inputs to these Models (three potential evapotranspiration [PET] grids and four soil water holding capacity [WHC] grids). The platform makes it possible to regionalise runoff flows, which are discretised in grids of mesh size 0.5 · 0.5. On the basis of these datasets and Models, we compare the performance of the various possible data-Model Combinations. The study area contains 49 catchments located in Cote d'Ivoire, Guinea, Mali, Burkina Faso and Niger. The analysis shows that the selected Models are hardly sensitive at all to the different PET grids but more sensitive to the soil grids. The GR2M Model clearly gives the best results for the set of catchments, but does not perform well over the entire study area, whatever the grid dataset used: It is thus seen to be difficult to define a single data-Model Combination that is optimal for simulating runoff in West Africa. As to the existence of a general' Model that could be used anywhere, we find that, for the selected study area, neither of the two Models selected fully meets the objective.
-
Simulation of runoff in West Africa: Is there a single data-Model Combination that produces the best simulation results?
Journal of Hydrology, 2008Co-Authors: A. Dezetter, S. Girard, J.e. Paturel, Gilles Mahe, S. Ardoin-bardin, E. ServatAbstract:Is there a single Combination of hydrologic Model and data that yields the best simulations of runoff from a set of catchments in West Africa in the coming decades? To answer this question, a semi-distributed Modelling platform was developed. It comprises two Models (GR2M and Water Balance Model) and various datasets as inputs to these Models (three potential evapotranspiration [PET] grids and four soil water holding capacity [WHC] grids). The platform makes it possible to regionalise runoff flows, which are discretised in grids of mesh size 0.5 degrees x 0.5 degrees. On the basis of these datasets and Models, we compare the performance of the various possible data-Model Combinations. The study area contains 49 catchments located in Cote d'lvoire, Guinea, Mali, Burkina Faso and Niger. The analysis shows that the selected Models are hardly sensitive at all to the different PET grids but more sensitive to the soil grids. The GR2M Model clearly gives the best results for the set of catchments, but does not perform well over the entire study area, whatever the grid dataset used: It is thus seen to be difficult to define a single data-Model Combination that is optimal for simulating runoff in West Africa. As to the existence of a "general" Model that could be used anywhere, we find that, for the selected study area, neither of the two Models selected fully meets the objective
Deepak Turaga - One of the best experts on this subject based on the ideXlab platform.
-
KDD - Class-distribution regularized consensus maximization for alleviating overfitting in Model Combination
Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining, 2014Co-Authors: Sihong Xie, Wei Fan, Jing Gao, Deepak TuragaAbstract:In data mining applications such as crowdsourcing and privacy-preserving data mining, one may wish to obtain consolidated predictions out of multiple Models without access to features of the data. Besides, multiple Models usually carry complementary predictive information, Model Combination can potentially provide more robust and accurate predictions by correcting independent errors from individual Models. Various methods have been proposed to combine predictions such that the final predictions are maximally agreed upon by multiple base Models. Though this maximum consensus principle has been shown to be successful, simply maximizing consensus can lead to less discriminative predictions and overfit the inevitable noise due to imperfect base Models. We argue that proper regularization for Model Combination approaches is needed to alleviate such overfitting effect. Specifically, we analyze the hypothesis spaces of several Model Combination methods and identify the trade-off between Model consensus and generalization ability. We propose a novel Model called Regularized Consensus Maximization (RCM), which is formulated as an optimization problem to combine the maximum consensus and large margin principles. We theoretically show that RCM has a smaller upper bound on generalization error compared to the version without regularization. Experiments show that the proposed algorithm outperforms a wide spectrum of state-of-the-art Model Combination methods on 11 tasks.
-
Class-distribution regularized consensus maximization for alleviating overfitting in Model Combination
Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '14, 2014Co-Authors: Sihong Xie, Deepak Turaga, Wei Fan, Jing Gao, Philip S YuAbstract:In data mining applications such as crowdsourcing and privacy-preserving data mining, one may wish to obtain consolidated predictions out of multiple Models without access to features of the data. Besides, multiple Models usually carry complementary predictive information, Model Combination can potentially provide more robust and accurate predictions by correcting independent errors from individual Models. Various methods have been proposed to combine predictions such that the final predictions are maximally agreed upon by multiple base Models. Though this maximum consensus principle has been shown to be successful, simply maximizing consensus can lead to less discriminative predictions and overfit the inevitable noise due to imperfect base Models. We argue that proper regularization for Model Combination approaches is needed to alleviate such overfitting effect. Specifically, we analyze the hypothesis spaces of several Model Combination methods and identify the trade-off between Model consensus and generalization ability. We propose a novel Model called Regularized Consensus Maximization (RCM), which is formulated as an optimization problem to combine the maximum consensus and large margin principles. We theoretically show that RCM has a smaller upper bound on generalization error compared to the version without regularization. Experiments show that the proposed algorithm outperforms a wide spectrum of state-of-the-art Model Combination methods on 11 tasks. © 2014 ACM.
Sihong Xie - One of the best experts on this subject based on the ideXlab platform.
-
KDD - Class-distribution regularized consensus maximization for alleviating overfitting in Model Combination
Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining, 2014Co-Authors: Sihong Xie, Wei Fan, Jing Gao, Deepak TuragaAbstract:In data mining applications such as crowdsourcing and privacy-preserving data mining, one may wish to obtain consolidated predictions out of multiple Models without access to features of the data. Besides, multiple Models usually carry complementary predictive information, Model Combination can potentially provide more robust and accurate predictions by correcting independent errors from individual Models. Various methods have been proposed to combine predictions such that the final predictions are maximally agreed upon by multiple base Models. Though this maximum consensus principle has been shown to be successful, simply maximizing consensus can lead to less discriminative predictions and overfit the inevitable noise due to imperfect base Models. We argue that proper regularization for Model Combination approaches is needed to alleviate such overfitting effect. Specifically, we analyze the hypothesis spaces of several Model Combination methods and identify the trade-off between Model consensus and generalization ability. We propose a novel Model called Regularized Consensus Maximization (RCM), which is formulated as an optimization problem to combine the maximum consensus and large margin principles. We theoretically show that RCM has a smaller upper bound on generalization error compared to the version without regularization. Experiments show that the proposed algorithm outperforms a wide spectrum of state-of-the-art Model Combination methods on 11 tasks.
-
Class-distribution regularized consensus maximization for alleviating overfitting in Model Combination
Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '14, 2014Co-Authors: Sihong Xie, Deepak Turaga, Wei Fan, Jing Gao, Philip S YuAbstract:In data mining applications such as crowdsourcing and privacy-preserving data mining, one may wish to obtain consolidated predictions out of multiple Models without access to features of the data. Besides, multiple Models usually carry complementary predictive information, Model Combination can potentially provide more robust and accurate predictions by correcting independent errors from individual Models. Various methods have been proposed to combine predictions such that the final predictions are maximally agreed upon by multiple base Models. Though this maximum consensus principle has been shown to be successful, simply maximizing consensus can lead to less discriminative predictions and overfit the inevitable noise due to imperfect base Models. We argue that proper regularization for Model Combination approaches is needed to alleviate such overfitting effect. Specifically, we analyze the hypothesis spaces of several Model Combination methods and identify the trade-off between Model consensus and generalization ability. We propose a novel Model called Regularized Consensus Maximization (RCM), which is formulated as an optimization problem to combine the maximum consensus and large margin principles. We theoretically show that RCM has a smaller upper bound on generalization error compared to the version without regularization. Experiments show that the proposed algorithm outperforms a wide spectrum of state-of-the-art Model Combination methods on 11 tasks. © 2014 ACM.