Browsing by Author "Lozano, J.A."
Now showing items 120 of 71

An active adaptation strategy for streaming time series classification based on elastic similarity measures
Oregi, I.; Pérez, A.; Del Ser, J.; Lozano, J.A. (20220521)In streaming time series classification problems, the goal is to predict the label associated to the most recently received observations over the stream according to a set of categorized reference patterns. In online ... 
AdHoc Explanation for Time Series Classification
Abanda, A.; Mori, U.; Lozano, J.A. (2022)In this work, a perturbationbased modelagnostic explanation method for time series classification is presented. One of the main novelties of the proposed method is that the considered perturbations are interpretable and ... 
Aggregated outputs by linear models: An application on marine litter beaching prediction
HernándezGonzález, J.; Inza, I.; Granado, I.; Basurko, O.C.; Fernández, J.A.; Lozano, J.A. (20190101)In regression, a predictive model which is able to anticipate the output of a new case is learnt from a set of previous examples. The output or response value of these examples used for model training is known. When learning ... 
Analysis of Dominant Classes in Universal Adversarial Perturbations
Vadillo, J.; Santana, R.; Lozano, J.A. (2022)The reasons why Deep Neural Networks are susceptible to being fooled by adversarial examples remains an open discussion. Indeed, many differ ent strategies can be employed to efficiently generate adversarial attacks, some ... 
Analysis of the sensitivity of the EndOfTurn Detection task to errors generated by the Automatic Speech Recognition process.
Montenegro, C.; Santana, R.; Lozano, J.A. (2021)An EndOfTurn Detection Module (EOTDM) is an essential component of au tomatic Spoken Dialogue Systems. The capability of correctly detecting whether a user’s utterance has ended or not improves the accuracy in interpreting ... 
Analyzing rare event, anomaly, novelty and outlier detection terms under the supervised classification framework
Carreño, A.; Inza, I.; Lozano, J.A. (20190901)In recent years, a variety of research areas have contributed to a set of related problems with rare event, anomaly, novelty and outlier detection terms as the main actors. These multiple research areas have created a ... 
Anatomy of the attraction basins: Breaking with the intuition
Hernando, L.; Mendiburu, A.; Lozano, J.A. (2019)olving combinatorial optimization problems efficiently requires the development of algorithms that consider the specific properties of the problems. In this sense, local search algorithms are designed over a neighborhood ... 
Are the artificially generated instances uniform in terms of difficulty?
Pérez, A.; Ceberio, J.; Lozano, J.A. (201806)In the field of evolutionary computation, it is usual to generate artificial benchmarks of instances that are used as a testbed to determine the performance of the algorithms at hand. In this context, a recent work on ... 
Bayesian inference for algorithm ranking analysis
Calvo, B.; Ceberio, J.; Lozano, J.A. (20180830)The statistical assessment of the empirical comparison of algorithms is an essential step in heuristic optimization. Classically, researchers have relied on the use of statistical tests. However, recently, concerns about ... 
Bayesian Optimization Approaches for Massively Multimodal Problems
Roman, I.; Mendiburu, A.; Santana, R.; Lozano, J.A. (2019)The optimization of massively multimodal functions is a challenging task, particularly for problems where the search space can lead the op timization process to local optima. While evolutionary algorithms have been ... 
Characterising the rankings produced by combinatorial optimisation problems and finding their intersections
Hernando, L.; Mendiburu, A.; Lozano, J.A. (2019)The aim of this paper is to introduce the concept of intersection between combinatorial optimisation problems. We take into account that most algorithms, in their machinery, do not consider the exact objective function ... 
A cheap feature selection approach for the K means algorithm
Capo, M.; Pérez, A.; Lozano, J.A. (202105)The increase in the number of features that need to be analyzed in a wide variety of areas, such as genome sequencing, computer vision or sensor networks, represents a challenge for the Kmeans algorithm. In this regard, ... 
Data generation approaches for topic classification in multilingual spoken dialog systems
Montenegro, C.; Santana, R.; Lozano, J.A. (2019)The conception of spokendialog systems (SDS) usually faces the problem of extending or adapting the system to multiple languages. This implies the creation of modules specically for the new languages, which is a time ... 
Delineation of site‐specific management zones using estimation of distribution algorithms
Lozano, J.A.; Velasco, J.; Vicencio, S.; CidGarcía, N.M. (2021)In this paper, we present a novel methodology to solve the problem of delineating homogeneous sitespecific management zones (SSMZ) in agricultural fields. This problem consists of dividing the field into small regions for ... 
Detection of Sand Dunes on Mars Using a Regular Vinebased Classification Approach
Carrera, D.; Bandeira, L.; Santana, R.; Lozano, J.A. (201808)This paper deals with the problem of detecting sand dunes from remotely sensed images of the surface of Mars. We build on previous approaches that propose methods to extract informative features for the classification of ... 
Distancebased exponential probability models on constrained combinatorial optimization problems
Ceberio, J.; Mendiburu, A.; Lozano, J.A. (20180830)Estimation of distribution algorithms have already demonstrated their utility when solving a broad range of combinatorial problems. However, there is still room for methodological improvements when approaching constrained ... 
Early classification of time series by simultaneously optimizing the accuracy and earliness
Mori, U.; Mendiburu, A.; Dasgupta, S.; Lozano, J.A. (201710)The problem of early classi cation of time series appears naturally in contexts where the data, of temporal nature, is collected over time, and early class predictions are interesting or even required. The objective is to ... 
Early classification of time series using multiobjective optimization techniques
Mori, U.; Mendiburu, A.; Miranda, I.M.; Lozano, J.A. (20190423)In early classification of time series the objective is to build models which are able to make classpredictions for time series as accurately and as early as possible, when only a part of the series is available. It is ... 
EDA++: Estimation of Distribution Algorithms with Feasibility Conserving Mechanisms for Constrained Continuous Optimization
Shirazi, A.; Ceberio, J.; Lozano, J.A. (20220225)Handling nonlinear constraints in continuous optimization is challenging, and finding a feasible solution is usually a difficult task. In the past few decades, various techniques have been developed to deal with linear ... 
Effects of reducing VMs management times on elastic applications
Pascual, J.A.; Lozano, J.A.; MiguelAlonso, J. (201805)Cloud infrastructures provide computing resources to applications in the form of Virtual Machines (VMs). Many applications deployed in cloud resources have an elastic behavior, that is, they change the number of servers ...