We applied genetic programming algorithm to learn the behavior of an occupant in single person office based on motion sensor data. The learned rules predict the presence and absence of the occupant with 80%–83% accuracy on testing data from 5 different offices. The rules indicate that the following variables may influence occupancy behavior: 1) the day of week, 2) the time of day, 3) the length of time the occupant spent in the previous state, 4) the length of time the occupant spent in the state prior to the previous state, 5) the length of time the occupant has been in the office since the first arrival of the day. We evaluate the rules with various statistics, which confirm some of the previous findings by other researchers. We also provide new insights about occupancy behavior of these offices that have not been reported previously.
{"title":"Modeling Occupancy Behavior for Energy Efficiency and Occupants Comfort Management in Intelligent Buildings","authors":"Tina Yu","doi":"10.1109/ICMLA.2010.111","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.111","url":null,"abstract":"We applied genetic programming algorithm to learn the behavior of an occupant in single person office based on motion sensor data. The learned rules predict the presence and absence of the occupant with 80%–83% accuracy on testing data from 5 different offices. The rules indicate that the following variables may influence occupancy behavior: 1) the day of week, 2) the time of day, 3) the length of time the occupant spent in the previous state, 4) the length of time the occupant spent in the state prior to the previous state, 5) the length of time the occupant has been in the office since the first arrival of the day. We evaluate the rules with various statistics, which confirm some of the previous findings by other researchers. We also provide new insights about occupancy behavior of these offices that have not been reported previously.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115542696","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}
Xavier Sierra-Canto, Francisco Madera-Ramirez, Víctor Uc Cetina
The Artificial Neural Networks (ANN) training represents a time-consuming process in machine learning systems. In this work we provide an implementation of the back-propagation algorithm on CUDA, a parallel computing architecture developed by NVIDIA. Using CUBLAS, a CUDA implementation of the Basic Linear Algebra Subprograms library (BLAS), the process is simplified, however, the use of kernels was necessary since CUBLAS does not have all the required operations. The implementation was tested with two standard benchmark data sets and the results show that the parallel training algorithm runs 63 times faster than its sequential version.
{"title":"Parallel Training of a Back-Propagation Neural Network Using CUDA","authors":"Xavier Sierra-Canto, Francisco Madera-Ramirez, Víctor Uc Cetina","doi":"10.1109/ICMLA.2010.52","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.52","url":null,"abstract":"The Artificial Neural Networks (ANN) training represents a time-consuming process in machine learning systems. In this work we provide an implementation of the back-propagation algorithm on CUDA, a parallel computing architecture developed by NVIDIA. Using CUBLAS, a CUDA implementation of the Basic Linear Algebra Subprograms library (BLAS), the process is simplified, however, the use of kernels was necessary since CUBLAS does not have all the required operations. The implementation was tested with two standard benchmark data sets and the results show that the parallel training algorithm runs 63 times faster than its sequential version.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117153141","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}
Kernel methods have been successfully used in many practical machine learning problems. Choosing a suitable kernel is left to the practitioner. A common way to an automatic selection of optimal kernels is to learn a linear combination of element kernels. In this paper, a novel framework of multiple kernel learning is proposed based on conditional entropy minimization criterion. For the proposed framework, three multiple kernel learning algorithms are derived. The algorithms are experimentally shown to be comparable to or outperform kernel Fisher discriminant analysis and other multiple kernel learning algorithms on benchmark data sets.
{"title":"Multiple Kernel Learning by Conditional Entropy Minimization","authors":"H. Hino, N. Reyhani, Noboru Murata","doi":"10.1109/ICMLA.2010.40","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.40","url":null,"abstract":"Kernel methods have been successfully used in many practical machine learning problems. Choosing a suitable kernel is left to the practitioner. A common way to an automatic selection of optimal kernels is to learn a linear combination of element kernels. In this paper, a novel framework of multiple kernel learning is proposed based on conditional entropy minimization criterion. For the proposed framework, three multiple kernel learning algorithms are derived. The algorithms are experimentally shown to be comparable to or outperform kernel Fisher discriminant analysis and other multiple kernel learning algorithms on benchmark data sets.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"70 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123472684","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}
Reinforcement learning algorithms that employ neural networks as function approximators have proven to be powerful tools for solving optimal control problems. However, their training and the validation of final policies can be cumbersome as neural networks can suffer from problems like local minima or over fitting. When using iterative methods, such as neural fitted Q-iteration, the problem becomes even more pronounced since the network has to be trained multiple times and the training process in one iteration builds on the network trained in the previous iteration. Therefore errors can accumulate. In this paper we propose to use ensembles of networks to make the learning process more robust and produce near-optimal policies more reliably. We name various ways of combining single networks to an ensemble that results in a final ensemble policy and show the potential of the approach using a benchmark application. Our experiments indicate that majority voting is superior to Q-averaging and using heterogeneous ensembles (different network topologies) is advisable.
{"title":"Ensembles of Neural Networks for Robust Reinforcement Learning","authors":"A. Hans, S. Udluft","doi":"10.1109/ICMLA.2010.66","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.66","url":null,"abstract":"Reinforcement learning algorithms that employ neural networks as function approximators have proven to be powerful tools for solving optimal control problems. However, their training and the validation of final policies can be cumbersome as neural networks can suffer from problems like local minima or over fitting. When using iterative methods, such as neural fitted Q-iteration, the problem becomes even more pronounced since the network has to be trained multiple times and the training process in one iteration builds on the network trained in the previous iteration. Therefore errors can accumulate. In this paper we propose to use ensembles of networks to make the learning process more robust and produce near-optimal policies more reliably. We name various ways of combining single networks to an ensemble that results in a final ensemble policy and show the potential of the approach using a benchmark application. Our experiments indicate that majority voting is superior to Q-averaging and using heterogeneous ensembles (different network topologies) is advisable.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124214893","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}
This paper proposes a Realized Trading Volatility (RTV) model for dynamically monitoring anomalous volatility in stock trading. Specifically, the RTV model first extracts the sequences for price volatility, volume volatility, and realized trading volatility. Then, the K-means algorithm is exploited for clustering the summary data of different stocks. The RTV model investigates the joint-volatility between share price and trading volume, and has the advantage of capturing anomalous trading volatility in a dynamic fashion. As a case study, we apply the RTV model for the analysis of real-world high-frequency stock data. For the resultant clusters, we focus on the categories with large volatility and study their statistical properties. Finally, we provide some empirical insights for the use of the RTV model.
{"title":"Clustering High-frequency Stock Data for Trading Volatility Analysis","authors":"Xiao-Wei Ai, Tianming Hu, Xi Li, Hui Xiong","doi":"10.1109/ICMLA.2010.56","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.56","url":null,"abstract":"This paper proposes a Realized Trading Volatility (RTV) model for dynamically monitoring anomalous volatility in stock trading. Specifically, the RTV model first extracts the sequences for price volatility, volume volatility, and realized trading volatility. Then, the K-means algorithm is exploited for clustering the summary data of different stocks. The RTV model investigates the joint-volatility between share price and trading volume, and has the advantage of capturing anomalous trading volatility in a dynamic fashion. As a case study, we apply the RTV model for the analysis of real-world high-frequency stock data. For the resultant clusters, we focus on the categories with large volatility and study their statistical properties. Finally, we provide some empirical insights for the use of the RTV model.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"1 9","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120899941","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}
Variable Ordering plays an important role when inducing Bayesian Networks. Previous works in the literature suggest that the use of genetic/evolutionary algorithms (EAs) for dealing with VO, when learning a Bayesian Network structure from data, is worth pursuing. This work proposes a new crossover operator, named Random Multi-point Crossover Operator (RMX), to be used with the Variable Ordering Evolutionary Algorithm (VOEA). Empirical results obtained by VOEA are compared to the ones achieved by VOGA (Variable Ordering Genetic Algorithm), and indicated improvement in the quality of VO and the induced BN structure.
{"title":"Evolutionary Algorithm Using Random Multi-point Crossover Operator for Learning Bayesian Network Structures","authors":"E. B. D. Santos, Estevam Hruschka, N. Ebecken","doi":"10.1109/ICMLA.2010.70","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.70","url":null,"abstract":"Variable Ordering plays an important role when inducing Bayesian Networks. Previous works in the literature suggest that the use of genetic/evolutionary algorithms (EAs) for dealing with VO, when learning a Bayesian Network structure from data, is worth pursuing. This work proposes a new crossover operator, named Random Multi-point Crossover Operator (RMX), to be used with the Variable Ordering Evolutionary Algorithm (VOEA). Empirical results obtained by VOEA are compared to the ones achieved by VOGA (Variable Ordering Genetic Algorithm), and indicated improvement in the quality of VO and the induced BN structure.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126522334","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}
The medium of music has evolved specifically for the expression of emotions, and it is natural for us to organize music in terms of its emotional associations. In previous work, we have modeled human response labels to music in the arousal-valence (A-V) representation of affect as a time-varying, stochastic distribution reflecting the ambiguous nature of the perception of mood. These distributions are used to predict A-V responses from acoustic features of the music alone via multi-variate regression. In this paper, we extend our framework to account for multiple regression mappings contingent upon a general location in A-V space. Furthermore, we model A-V state as the latent variable of a linear dynamical system, more explicitly capturing the dynamics of musical mood. We validate this extension using a "genie-bounded" approach, in which we assume that a piece of music is correctly clustered in A-V space a priori, demonstrating significantly higher theoretical performance than the previous single-regressor approach.
{"title":"Prediction of Time-Varying Musical Mood Distributions Using Kalman Filtering","authors":"Erik M. Schmidt, Youngmoo E. Kim","doi":"10.1109/ICMLA.2010.101","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.101","url":null,"abstract":"The medium of music has evolved specifically for the expression of emotions, and it is natural for us to organize music in terms of its emotional associations. In previous work, we have modeled human response labels to music in the arousal-valence (A-V) representation of affect as a time-varying, stochastic distribution reflecting the ambiguous nature of the perception of mood. These distributions are used to predict A-V responses from acoustic features of the music alone via multi-variate regression. In this paper, we extend our framework to account for multiple regression mappings contingent upon a general location in A-V space. Furthermore, we model A-V state as the latent variable of a linear dynamical system, more explicitly capturing the dynamics of musical mood. We validate this extension using a \"genie-bounded\" approach, in which we assume that a piece of music is correctly clustered in A-V space a priori, demonstrating significantly higher theoretical performance than the previous single-regressor approach.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127809532","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 this paper we propose a novel approach for introducing semantic relations into the bag-of-words framework for recognizing human actions. We represent visual words in two different views: the original features and the document co-occurrence representation. The latter view conveys semantic relations but is large, sparse and noisy. We use canonical correlation analysis between the two views to find a subspace in which the words are more semantically distributed. We apply k-means clustering in the computed space to find semantically meaningful clusters and use them as the semantic visual vocabulary. Incorporating the semantic visual vocabulary the features are quantized to form more discriminative histograms. Eventually the histograms are classified using an SVM classifier. We have tested our approach on KTH action dataset and achieved promising results.
{"title":"Multi-view Clustering of Visual Words Using Canonical Correlation Analysis for Human Action Recognition","authors":"Behrouz Saghafi, D. Rajan","doi":"10.1109/ICMLA.2010.102","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.102","url":null,"abstract":"In this paper we propose a novel approach for introducing semantic relations into the bag-of-words framework for recognizing human actions. We represent visual words in two different views: the original features and the document co-occurrence representation. The latter view conveys semantic relations but is large, sparse and noisy. We use canonical correlation analysis between the two views to find a subspace in which the words are more semantically distributed. We apply k-means clustering in the computed space to find semantically meaningful clusters and use them as the semantic visual vocabulary. Incorporating the semantic visual vocabulary the features are quantized to form more discriminative histograms. Eventually the histograms are classified using an SVM classifier. We have tested our approach on KTH action dataset and achieved promising results.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116834150","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 this paper, we describe a framework for solving computationally hard, distributed function optimization problems using reinforcement learning techniques. In particular, we model a function optimization problem as an identical payoff game played by a team of reinforcement learning agents. The team performs a stochastic search through the domain space of the parameters of the function. However, current game learning algorithms suffer from significant memory requirement, significant communication overhead and slow convergence. To alleviate these problems, we present novel decentralized and partially decentralized reinforcement learning algorithms for the team. Simulation results are presented for the NP-Hard sensor subset selection problem to show that the agents learn locally optimal parameter values and illustrate the advantages of the proposed algorithms.
{"title":"Decentralized and Partially Decentralized Reinforcement Learning for Distributed Combinatorial Optimization Problems","authors":"Omkar J. Tilak, S. Mukhopadhyay","doi":"10.1109/ICMLA.2010.64","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.64","url":null,"abstract":"In this paper, we describe a framework for solving computationally hard, distributed function optimization problems using reinforcement learning techniques. In particular, we model a function optimization problem as an identical payoff game played by a team of reinforcement learning agents. The team performs a stochastic search through the domain space of the parameters of the function. However, current game learning algorithms suffer from significant memory requirement, significant communication overhead and slow convergence. To alleviate these problems, we present novel decentralized and partially decentralized reinforcement learning algorithms for the team. Simulation results are presented for the NP-Hard sensor subset selection problem to show that the agents learn locally optimal parameter values and illustrate the advantages of the proposed algorithms.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132987696","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}
F. Lin, Jonq-Chin Hwang, K. Tan, Zong-Han Lu, Yung-Ruei Chang
An intelligent control stand-alone doubly-fed induction generator (DFIG) system using proportional-integral-derivative neural network (PIDNN) is proposed in this study. This system can be applied as a stand-alone power supply system or as the emergency power system when the electricity grid fails for all sub-synchronous, synchronous and super-synchronous conditions. The rotor side converter is controlled using the field-oriented control to produce three-phase stator voltages with constant magnitude and frequency at different rotor speeds. Moreover, the stator side converter, which is also controlled using field-oriented control, is primarily implemented to maintain the magnitude of the DC-link voltage. Furthermore, the intelligent PIDNN controller is proposed for both the rotor and stator side converters to improve the transient and steady-state responses of the DFIG system for different operating conditions. Both the network structure and on-line learning algorithm are introduced in detail. Finally, the feasibility of the proposed control scheme is verified through experimentation.
{"title":"Control of Doubly-Fed Induction Generator System Using PIDNNs","authors":"F. Lin, Jonq-Chin Hwang, K. Tan, Zong-Han Lu, Yung-Ruei Chang","doi":"10.1109/ICMLA.2010.104","DOIUrl":"https://doi.org/10.1109/ICMLA.2010.104","url":null,"abstract":"An intelligent control stand-alone doubly-fed induction generator (DFIG) system using proportional-integral-derivative neural network (PIDNN) is proposed in this study. This system can be applied as a stand-alone power supply system or as the emergency power system when the electricity grid fails for all sub-synchronous, synchronous and super-synchronous conditions. The rotor side converter is controlled using the field-oriented control to produce three-phase stator voltages with constant magnitude and frequency at different rotor speeds. Moreover, the stator side converter, which is also controlled using field-oriented control, is primarily implemented to maintain the magnitude of the DC-link voltage. Furthermore, the intelligent PIDNN controller is proposed for both the rotor and stator side converters to improve the transient and steady-state responses of the DFIG system for different operating conditions. Both the network structure and on-line learning algorithm are introduced in detail. Finally, the feasibility of the proposed control scheme is verified through experimentation.","PeriodicalId":336514,"journal":{"name":"2010 Ninth International Conference on Machine Learning and Applications","volume":"603 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133466819","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：