Ponències/Comunicacions de congressos
http://hdl.handle.net/2117/3689
2017-05-29T09:43:28ZBayesian semi non-negative matrix factorisation
http://hdl.handle.net/2117/103878
Bayesian semi non-negative matrix factorisation
Vilamala Muñoz, Albert; Vellido Alcacena, Alfredo; Belanche Muñoz, Luis Antonio
Non-negative Matrix Factorisation (NMF) has become a standard method for source identification when data, sources and mixing coefficients are constrained to be positive-valued. The method has recently been extended to allow for negative-valued data and sources in the form of Semi-and Convex-NMF. In this paper, we re-elaborate Semi-NMF within a full Bayesian framework. This provides solid foundations for parameter estimation and, importantly, a principled method to address the problem of choosing the most adequate number of sources to describe the observed data. The proposed Bayesian Semi-NMF is preliminarily evaluated here in a real neuro-oncology problem.
2017-05-02T08:11:19ZVilamala Muñoz, AlbertVellido Alcacena, AlfredoBelanche Muñoz, Luis AntonioNon-negative Matrix Factorisation (NMF) has become a standard method for source identification when data, sources and mixing coefficients are constrained to be positive-valued. The method has recently been extended to allow for negative-valued data and sources in the form of Semi-and Convex-NMF. In this paper, we re-elaborate Semi-NMF within a full Bayesian framework. This provides solid foundations for parameter estimation and, importantly, a principled method to address the problem of choosing the most adequate number of sources to describe the observed data. The proposed Bayesian Semi-NMF is preliminarily evaluated here in a real neuro-oncology problem.A methodological approach for algorithmic composition systems' parameter spaces aesthetic exploration
http://hdl.handle.net/2117/102221
A methodological approach for algorithmic composition systems' parameter spaces aesthetic exploration
Paz Ortiz, Iván; Nebot Castells, M. Àngela; Romero Merino, Enrique; Múgica Álvarez, Francisco; Vellido Alcacena, Alfredo
Algorithmic composition is the process of creating musical material by means of formal methods. As a consequence of its design, algorithmic composition systems are (explicitly or implicitly) described in terms of parameters. Thus, parameter space exploration plays a key role in learning the system's capabilities. However, in the computer music field, this task has received little attention. This is due in part, because the produced changes on the human perception of the outputs, as a response to changes on the parameters, could be highly nonlinear, therefore models with strongly predictable outputs are needed. The present work describes a methodology for the human perceptual (or aesthetic) exploration of generative systems' parameter spaces. As the systems' outputs are intended to produce an aesthetic experience on humans, audition plays a central role in the process. The methodology starts from a set of parameter combinations which are perceptually evaluated by the user. The sampling process of such combinations depends on the system under study and possible on heuristic considerations. The evaluated set is processed by a compaction algorithm able to generate linguistic rules describing the distinct perceptions (classes) of the user evaluation. The semantic level of the extracted rules allows for interpretability, while showing great potential in describing high and low-level musical entities. As the resulting rules represent discrete points in the parameter space, further possible extensions for interpolation between points are also discussed. Finally, some practical implementations and paths for further research are presented.
2017-03-09T15:00:28ZPaz Ortiz, IvánNebot Castells, M. ÀngelaRomero Merino, EnriqueMúgica Álvarez, FranciscoVellido Alcacena, AlfredoAlgorithmic composition is the process of creating musical material by means of formal methods. As a consequence of its design, algorithmic composition systems are (explicitly or implicitly) described in terms of parameters. Thus, parameter space exploration plays a key role in learning the system's capabilities. However, in the computer music field, this task has received little attention. This is due in part, because the produced changes on the human perception of the outputs, as a response to changes on the parameters, could be highly nonlinear, therefore models with strongly predictable outputs are needed. The present work describes a methodology for the human perceptual (or aesthetic) exploration of generative systems' parameter spaces. As the systems' outputs are intended to produce an aesthetic experience on humans, audition plays a central role in the process. The methodology starts from a set of parameter combinations which are perceptually evaluated by the user. The sampling process of such combinations depends on the system under study and possible on heuristic considerations. The evaluated set is processed by a compaction algorithm able to generate linguistic rules describing the distinct perceptions (classes) of the user evaluation. The semantic level of the extracted rules allows for interpretability, while showing great potential in describing high and low-level musical entities. As the resulting rules represent discrete points in the parameter space, further possible extensions for interpolation between points are also discussed. Finally, some practical implementations and paths for further research are presented.A decision making support tool: The resilience management fuzzy controller
http://hdl.handle.net/2117/102219
A decision making support tool: The resilience management fuzzy controller
González Cardenas, Rubén; Nebot Castells, M. Àngela; Múgica Álvarez, Francisco; Vellido Alcacena, Alfredo
In this paper a fuzzy controller capable to perform an automated estimation of the period of time necessary to recover a resilience level is proposed. Estimations where made by considering realistic time-dependent action changes for a set of resilience indicators originally proposed by Cardona (2001) and modified by Cardenas et al (2015). The fuzzy resilience controller works using two output control variables and four input variables designed to resemble politics decisions made over resilience recovery while considering an economical national growth factor. We applied the fuzzy controller onto Barcelona Spain, where different recovery times where estimated in terms of variations in Spaniard GDP (Gross domestic product) inter anual rate of change. This Decision Support System might be helpful to assist disaster reduction planning by allowing decision takers, governs or institutions to achieve reliable recovery time estimations while a proper supervision and control of resilience indicators progress is performed and an open evaluation and scrutiny of applied policies is made.
2017-03-09T14:48:13ZGonzález Cardenas, RubénNebot Castells, M. ÀngelaMúgica Álvarez, FranciscoVellido Alcacena, AlfredoIn this paper a fuzzy controller capable to perform an automated estimation of the period of time necessary to recover a resilience level is proposed. Estimations where made by considering realistic time-dependent action changes for a set of resilience indicators originally proposed by Cardona (2001) and modified by Cardenas et al (2015). The fuzzy resilience controller works using two output control variables and four input variables designed to resemble politics decisions made over resilience recovery while considering an economical national growth factor. We applied the fuzzy controller onto Barcelona Spain, where different recovery times where estimated in terms of variations in Spaniard GDP (Gross domestic product) inter anual rate of change. This Decision Support System might be helpful to assist disaster reduction planning by allowing decision takers, governs or institutions to achieve reliable recovery time estimations while a proper supervision and control of resilience indicators progress is performed and an open evaluation and scrutiny of applied policies is made.Multivariate dynamic kernels for financial time series forecasting
http://hdl.handle.net/2117/102167
Multivariate dynamic kernels for financial time series forecasting
Peña, Mauricio; Arratia Quesada, Argimiro Alejandro; Belanche Muñoz, Luis Antonio
We propose a forecasting procedure based on multivariate dynamic kernels, with the capability of integrating information measured at different frequencies and at irregular time intervals in financial markets. A data compression process redefines the original financial time series into temporal data blocks, analyzing the temporal information of multiple time intervals. The analysis is done through multivariate dynamic kernels within support vector regression. We also propose two kernels for financial time series that are computationally efficient without a sacrifice on accuracy. The efficacy of the methodology is demonstrated by empirical experiments on forecasting the challenging S&P500 market.
The final publication is available at http://link.springer.com/chapter/10.1007/978-3-319-44781-0_40
2017-03-09T08:50:56ZPeña, MauricioArratia Quesada, Argimiro AlejandroBelanche Muñoz, Luis AntonioWe propose a forecasting procedure based on multivariate dynamic kernels, with the capability of integrating information measured at different frequencies and at irregular time intervals in financial markets. A data compression process redefines the original financial time series into temporal data blocks, analyzing the temporal information of multiple time intervals. The analysis is done through multivariate dynamic kernels within support vector regression. We also propose two kernels for financial time series that are computationally efficient without a sacrifice on accuracy. The efficacy of the methodology is demonstrated by empirical experiments on forecasting the challenging S&P500 market.Automated quality control for proton magnetic resonance spectroscopy data using convex non-negative matrix factorization
http://hdl.handle.net/2117/99395
Automated quality control for proton magnetic resonance spectroscopy data using convex non-negative matrix factorization
Mocioiu, Victor; Kyathanahally, Sreenath P.; Arús, Carles; Vellido Alcacena, Alfredo; Julià Sapé, Margarida
Proton Magnetic Resonance Spectroscopy (1H MRS) has proven its diagnostic potential in a variety of conditions. However, MRS is not yet widely used in clinical routine because of the lack of experts on its diagnostic interpretation. Although data-based decision support systems exist to aid diagnosis, they often take for granted that the
data is of good quality, which is not always the case in a real application context. Systems based on models built with bad quality data are likely to underperform in their decision support tasks. In this study, we propose a system to filter out such bad quality data. It is based on convex Non-Negative Matrix Factorization models, used as a dimensionality reduction procedure, and on the use of several classifiers to discriminate between good and bad quality data.
2017-01-17T08:51:09ZMocioiu, VictorKyathanahally, Sreenath P.Arús, CarlesVellido Alcacena, AlfredoJulià Sapé, MargaridaProton Magnetic Resonance Spectroscopy (1H MRS) has proven its diagnostic potential in a variety of conditions. However, MRS is not yet widely used in clinical routine because of the lack of experts on its diagnostic interpretation. Although data-based decision support systems exist to aid diagnosis, they often take for granted that the
data is of good quality, which is not always the case in a real application context. Systems based on models built with bad quality data are likely to underperform in their decision support tasks. In this study, we propose a system to filter out such bad quality data. It is based on convex Non-Negative Matrix Factorization models, used as a dimensionality reduction procedure, and on the use of several classifiers to discriminate between good and bad quality data.A machine learning pipeline for supporting differentiation of glioblastomas from single brain metastases
http://hdl.handle.net/2117/97584
A machine learning pipeline for supporting differentiation of glioblastomas from single brain metastases
Mocioiu, Victor; de Barros, Nuno M. Pedrosa; Ortega Martorell, Sandra; Slotboom, Johannes; Knecht, Urspeter; Arús, Carles; Vellido Alcacena, Alfredo; Julià Sapé, Margarida
Machine learning has provided, over the last decades, tools for knowledge extraction in complex medical domains. Most of these tools, though, are ad hoc solutions and lack the systematic approach that would be required to become mainstream in medical practice. In this brief paper, we define a machine learning-based analysis pipeline for helping in a difficult problem in the field of neuro-oncology, namely the discrimination of brain glioblastomas from single brain metastases. This pipeline involves source extraction using k-Meansinitialized Convex Non-negative Matrix Factorization and a collection of classifiers, including Logistic Regression, Linear Discriminant Analysis, AdaBoost, and Random Forests.
2016-12-01T10:29:18ZMocioiu, Victorde Barros, Nuno M. PedrosaOrtega Martorell, SandraSlotboom, JohannesKnecht, UrspeterArús, CarlesVellido Alcacena, AlfredoJulià Sapé, MargaridaMachine learning has provided, over the last decades, tools for knowledge extraction in complex medical domains. Most of these tools, though, are ad hoc solutions and lack the systematic approach that would be required to become mainstream in medical practice. In this brief paper, we define a machine learning-based analysis pipeline for helping in a difficult problem in the field of neuro-oncology, namely the discrimination of brain glioblastomas from single brain metastases. This pipeline involves source extraction using k-Meansinitialized Convex Non-negative Matrix Factorization and a collection of classifiers, including Logistic Regression, Linear Discriminant Analysis, AdaBoost, and Random Forests.Instance and feature weighted k-nearest-neighbors algorithm
http://hdl.handle.net/2117/97582
Instance and feature weighted k-nearest-neighbors algorithm
Prat, Gabriel; Belanche Muñoz, Luis Antonio
We present a novel method that aims at providing a more stable selection of feature subsets when variations in the training process occur. This is accomplished by using an instance-weighting process -assigning different importances to instances as a preprocessing step to a feature weighting method that is independent of the learner, and then making good use of both sets of computed weigths in a standard Nearest-Neighbours classifier.
We report extensive experimentation in well-known benchmarking datasets as well as some challenging microarray
gene expression problems. Our results show increases in stability for most subset sizes and most problems, without
compromising prediction accuracy.
2016-12-01T10:15:32ZPrat, GabrielBelanche Muñoz, Luis AntonioWe present a novel method that aims at providing a more stable selection of feature subsets when variations in the training process occur. This is accomplished by using an instance-weighting process -assigning different importances to instances as a preprocessing step to a feature weighting method that is independent of the learner, and then making good use of both sets of computed weigths in a standard Nearest-Neighbours classifier.
We report extensive experimentation in well-known benchmarking datasets as well as some challenging microarray
gene expression problems. Our results show increases in stability for most subset sizes and most problems, without
compromising prediction accuracy.Physics and machine learning: Emerging paradigms
http://hdl.handle.net/2117/97581
Physics and machine learning: Emerging paradigms
Martín Guerrero, José; Lisboa, Paulo J G; Vellido Alcacena, Alfredo
Current research in Machine Learning (ML) combines the study of variations on well-established methods with cutting-edge breakthroughs based on completely new approaches. Among the latter, emerging paradigms from Physics have taken special relevance in recent years. Although still in its initial stages, Quantum Machine Learning (QML) shows promising ways to speed up some of the costly ML calculations with a similar or even better performance than existing approaches. Two additional advantages are related to the intrinsic probabilistic approach of QML, since quantum states are genuinely probabilistic, and to the capability of finding the global optimum of a given cost function by means of adiabatic quantum optimization, thus circumventing the usual problem of local minima. Another Physics approach for ML comes from Statistical Physics and is linked to Information theory in supervised and semi-supervised learning frameworks. On the other hand, and from the perspective of Physics, ML can provide solutions by extracting knowledge from huge amounts of data, as it is common in many experiments in the field, such as those related to High Energy Physics for elementary-particle research and Observational Astronomy.
2016-12-01T10:08:53ZMartín Guerrero, JoséLisboa, Paulo J GVellido Alcacena, AlfredoCurrent research in Machine Learning (ML) combines the study of variations on well-established methods with cutting-edge breakthroughs based on completely new approaches. Among the latter, emerging paradigms from Physics have taken special relevance in recent years. Although still in its initial stages, Quantum Machine Learning (QML) shows promising ways to speed up some of the costly ML calculations with a similar or even better performance than existing approaches. Two additional advantages are related to the intrinsic probabilistic approach of QML, since quantum states are genuinely probabilistic, and to the capability of finding the global optimum of a given cost function by means of adiabatic quantum optimization, thus circumventing the usual problem of local minima. Another Physics approach for ML comes from Statistical Physics and is linked to Information theory in supervised and semi-supervised learning frameworks. On the other hand, and from the perspective of Physics, ML can provide solutions by extracting knowledge from huge amounts of data, as it is common in many experiments in the field, such as those related to High Energy Physics for elementary-particle research and Observational Astronomy.A proposal for climate change resilience management through fuzzy controllers
http://hdl.handle.net/2117/90715
A proposal for climate change resilience management through fuzzy controllers
González Cárdenas, Rubén; Nebot Castells, M. Àngela; Múgica Álvarez, Francisco
We aim towards the implementation of a set of fuzzy controllers capable to perform automated estimation of the period of time necessary to recover a resilience level through the non-linear influence of a set of interrelated climate change resilience indicators constrained by social-based variables. This fuzzy controller set, working together with a fuzzy inference system type Mamdani, will be capable to estimate the proper adjustments to be done onto system’s elements in order to achieve a certain resilience level, while a general estimation of required costs is appraised. The final tool can then be used to provide guidelines for strategic vulnerability planning and monitoring through a clear understanding between investments and results, while an open evaluation and scrutiny of applied policies is made. In this paper the main strategy to achieve the mentioned objectives is
presented and discussed.
2016-10-13T07:55:07ZGonzález Cárdenas, RubénNebot Castells, M. ÀngelaMúgica Álvarez, FranciscoWe aim towards the implementation of a set of fuzzy controllers capable to perform automated estimation of the period of time necessary to recover a resilience level through the non-linear influence of a set of interrelated climate change resilience indicators constrained by social-based variables. This fuzzy controller set, working together with a fuzzy inference system type Mamdani, will be capable to estimate the proper adjustments to be done onto system’s elements in order to achieve a certain resilience level, while a general estimation of required costs is appraised. The final tool can then be used to provide guidelines for strategic vulnerability planning and monitoring through a clear understanding between investments and results, while an open evaluation and scrutiny of applied policies is made. In this paper the main strategy to achieve the mentioned objectives is
presented and discussed.Probability ridges and distortion flows: Visualizing multivariate time series using a variational Bayesian manifold learning method
http://hdl.handle.net/2117/82956
Probability ridges and distortion flows: Visualizing multivariate time series using a variational Bayesian manifold learning method
Tosi, Alessandra; Olier, Iván; Vellido Alcacena, Alfredo
Time-dependent natural phenomena and artificial processes can often be quantitatively expressed as multivariate time series (MTS). As in any other process of knowledge extraction from data, the analyst can benefit from the exploration of the characteristics of MTS through data visualization. This visualization often becomes difficult to interpret when MTS are modelled using nonlinear techniques. Despite their flexibility, nonlinear models can be rendered useless if such interpretability is lacking. In this brief paper, we model MTS using Variational Bayesian Generative Topographic Mapping Through Time (VB-GTM-TT), a variational Bayesian variant of a constrained hidden Markov model of the manifold learning family defined for MTS visualization. We aim to increase its interpretability by taking advantage of two results of the probabilistic definition of the model: the explicit estimation of probabilities of transition between states described in the visualization space and the quantification of the nonlinear mapping distortion.
2016-02-15T15:20:06ZTosi, AlessandraOlier, IvánVellido Alcacena, AlfredoTime-dependent natural phenomena and artificial processes can often be quantitatively expressed as multivariate time series (MTS). As in any other process of knowledge extraction from data, the analyst can benefit from the exploration of the characteristics of MTS through data visualization. This visualization often becomes difficult to interpret when MTS are modelled using nonlinear techniques. Despite their flexibility, nonlinear models can be rendered useless if such interpretability is lacking. In this brief paper, we model MTS using Variational Bayesian Generative Topographic Mapping Through Time (VB-GTM-TT), a variational Bayesian variant of a constrained hidden Markov model of the manifold learning family defined for MTS visualization. We aim to increase its interpretability by taking advantage of two results of the probabilistic definition of the model: the explicit estimation of probabilities of transition between states described in the visualization space and the quantification of the nonlinear mapping distortion.