TECNALIA Publications Repository :: Browsing by Author "Lobo, Jesus L."

Browsing by Author "Lobo, Jesus L."

Now showing 1 - 20 of 36

Adaptive long-term traffic state estimation with evolving spiking neural networks
(2019-04) Laña, Ibai; Lobo, Jesus L.; Capecci, Elisa; Del Ser, Javier; Kasabov, Nikola; IA
Due to the nature of traffic itself, most traffic forecasting models reported in literature aim at producing short-term predictions, yet their performance degrades when the prediction horizon is increased. The scarce long-term estimation strategies currently found in the literature are commonly based on the detection and assignment to patterns, but their performance decays when unexpected events provoke non predictable changes, or if the allocation to a traffic pattern is inaccurate. This work introduces a method to obtain long-term pattern forecasts and adapt them to real-time circumstances. To this end, a long-term estimation scheme based on the automated discovery of patterns is proposed and integrated with an on-line change detection and adaptation mechanism. The framework takes advantage of the architecture of evolving Spiking Neural Networks (eSNN) to perform adaptations without retraining the model, allowing the whole system to work autonomously in an on-line fashion. Its performance is assessed over a real scenario with 5 min data of a 6-month span of traffic in the center of Madrid, Spain. Significant accuracy gains are obtained when applying the proposed on-line adaptation mechanism on days with special, non-predictable events that degrade the quality of their long-term traffic forecasts.
AT-MFCGA: An Adaptive Transfer-guided Multifactorial Cellular Genetic Algorithm for Evolutionary Multitasking
(2021-09) Osaba, Eneko; Del Ser, Javier; Martinez, Aritz D.; Lobo, Jesus L.; Herrera, Francisco; Quantum; IA
Transfer Optimization is an incipient research area dedicated to solving multiple optimization tasks simultaneously. Among the different approaches that can address this problem effectively, Evolutionary Multitasking resorts to concepts from Evolutionary Computation to solve multiple problems within a single search process. In this paper we introduce a novel adaptive metaheuristic algorithm to deal with Evolutionary Multitasking environments coined as Adaptive Transfer-guided Multifactorial Cellular Genetic Algorithm (AT-MFCGA). AT-MFCGA relies on cellular automata to implement mechanisms in order to exchange knowledge among the optimization problems under consideration. Furthermore, our approach is able to explain by itself the synergies among tasks that were encountered and exploited during the search, which helps us to understand interactions between related optimization tasks. A comprehensive experimental setup is designed to assess and compare the performance of AT-MFCGA to that of other renowned Evolutionary Multitasking alternatives (MFEA and MFEA-II). Experiments comprise 11 multitasking scenarios composed of 20 instances of 4 combinatorial optimization problems, yielding the largest discrete multitasking environment solved to date. Results are conclusive in regard to the superior quality of solutions provided by AT-MFCGA with respect to the rest of the methods, which are complemented by a quantitative examination of the genetic transferability among tasks throughout the search process.
Cognitive workload classification using eye-tracking and EEG data
(Association for Computing Machinery, Inc, 2016-09-14) Lobo, Jesus L.; Ser, Javier Del; De Simone, Flavia; Presta, Roberta; Collina, Simona; Moravek, Zdenek; IA
It has been shown that an increased mental workload in pilots could lead to a decrease in their situation awareness, which could lead, in turn, to a worse piloting performance and ultimately to critical human errors. Assessing the current pilot's psycho-physiological state is a hot topic of interest for developing advanced embedded cockpits systems capable of adapting their behavior to the state and performance of the pilot. In this work, we investigate a method to classify different levels of cognitive workload starting from synchronized EEG and eye-tracking information. The classifier object of the research is targeted to score a performance high enough to be applicable as a gauge for performance of unobtrusive monitoring systems working with data of lower quality.
Community detection in graphs based on surprise maximization using firefly heuristics
(Institute of Electrical and Electronics Engineers Inc., 2016-11-14) Del Ser, Javier; Lobo, Jesus L.; Villar-Rodriguez, Esther; Bilbao, Miren Nekane; Perfecto, Cristina; IA; Quantum
The detection of node clusters (communities) in graphs has been at the core of many modeling paradigms emerging in different fields and disciplines such as Social Sciences, Biology, Chemistry, Telecommunications and Linguistics. When evaluating the quality of a clustering arrangement unsupervised metrics can be utilized (e.g. modularity), which all rely on structural and topological characteristics of the cluster space rather than on an observed ground of truth that should be achieved. One of such metrics is the recently published Surprise, which evaluates how statistically unlikely a given clustering arrangement is with respect to a random network featuring the same distribution of nodes per cluster. To maximize this metric, a number of algorithms have been proposed in the literature, but their comparative performance varies significantly between networks of different shape and size. In this article a novel heuristic community detection approach is proposed as a means to achieve a universally well-performing tool for graph clustering based on Surprise maximization. The heuristic scheme relies on the search procedure of the so-called Firefly Algorithm, a nature-inspired meta-heuristic solver based on the collective behavior, mutual attractiveness and random yet controlled movement of these insects. The proposed technique emulates these observed behavioral patterns of fireflies in the genotype of the graph clustering problem rather than on an encoded representation of its search space (phenotype). Simulation results evince that the performance of our community detection scheme generalizes better than other schemes when applied over synthetically generated graphs with varying properties.
Concept tracking and adaptation for drifting data streams under extreme verification latency
(Springer Verlag, 2018) Arostegi, Maria; Torre-Bastida, Ana I.; Lobo, Jesus L.; Bilbao, Miren Nekane; Del Ser, Javier; IA; Tecnalia Research & Innovation; HPA
When analyzing large-scale streaming data towards resolving classification problems, it is often assumed that true labels of the incoming data are available right after being predicted. This assumption allows online learning models to efficiently detect and accommodate non-stationarities in the distribution of the arriving data (concept drift). However, this assumption does not hold in many practical scenarios where a delay exists between predicted and class labels, to the point of lacking this supervision for an infinite period of time (extreme verification latency). In this case, the development of learning algorithms capable of adapting to drifting environments without any external supervision remains a challenging research area to date. In this context, this work proposes a simple yet effective learning technique to classify non-stationary data streams under extreme verification latency. The intuition motivating the design of our technique is to predict the trajectory of concepts in the feature space. The estimation of the region where concepts may reside in the future can be then exploited for producing more updated predictions for newly arriving examples, ultimately enhancing its accuracy during this unsupervised drifting period. Our approach is compared to a benchmark of incremental and static learning methods over a set of public non-stationary synthetic datasets. Results obtained by our passive learning method are promising and encourage further research aimed at generalizing its applicability to other types of drifts.
CURIE: a cellular automaton for concept drift detection: a cellular automaton for concept drift detection
(2021-11) Lobo, Jesus L.; Del Ser, Javier; Osaba, Eneko; Bifet, Albert; Herrera, Francisco; IA; Quantum
Data stream mining extracts information from large quantities of data flowing fast and continuously (data streams). They are usually affected by changes in the data distribution, giving rise to a phenomenon referred to as concept drift. Thus, learning models must detect and adapt to such changes, so as to exhibit a good predictive performance after a drift has occurred. In this regard, the development of effective drift detection algorithms becomes a key factor in data stream mining. In this work we propose CURIECURIE, a drift detector relying on cellular automata. Specifically, in CURIECURIE the distribution of the data stream is represented in the grid of a cellular automata, whose neighborhood rule can then be utilized to detect possible distribution changes over the stream. Computer simulations are presented and discussed to show that CURIECURIE, when hybridized with other base learners, renders a competitive behavior in terms of detection metrics and classification accuracy. CURIECURIE is compared with well-established drift detectors over synthetic datasets with varying drift characteristics.
Deep Echo State Networks for Short-Term Traffic Forecasting: Performance Comparison and Statistical Assessment
(Institute of Electrical and Electronics Engineers Inc., 2020-09-20) Del Ser, Javier; Laña, Ibai; Manibardo, Eric L.; Oregi, Izaskun; Osaba, Eneko; Lobo, Jesus L.; Bilbao, Miren Nekane; Vlahogianni, Eleni I.; IA; Quantum
In short-term traffic forecasting, the goal is to accurately predict future values of a traffic parameter of interest occurring shortly after the prediction is queried. The activity reported in this long-standing research field has been lately dominated by different Deep Learning approaches, yielding overly complex forecasting models that in general achieve accuracy gains of questionable practical utility. In this work we elaborate on the performance of Deep Echo State Networks for this particular task. The efficient learning algorithm and simpler parametric configuration of these alternative modeling approaches make them emerge as a competitive traffic forecasting method for real ITS applications deployed in devices and systems with stringently limited computational resources. An extensive comparison benchmark is designed with real traffic data captured over the city of Madrid (Spain), amounting to more than 130 Automatic Traffic Readers (ATRs) and several shallow learning, ensembles and Deep Learning models. Results from this comparison benchmark and the analysis of the statistical significance of the reported performance gaps are decisive: Deep Echo State Networks achieve more accurate traffic forecasts than the rest of considered modeling counterparts.
Drift detection over non-stationary data streams using evolving spiking neural networks
(Springer Verlag, 2018) Lobo, Jesus L.; Del Ser, Javier; Laña, Ibai; Bilbao, Miren Nekane; Kasabov, Nikola; IA
Drift detection in changing environments is a key factor for those active adaptive methods which require trigger mechanisms for drift adaptation. Most approaches are relied on a base learner that provides accuracies or error rates to be analyzed by an algorithm. In this work we propose the use of evolving spiking neural networks as a new form of drift detection, which resorts to the own architectural changes of this particular class of models to estimate the drift location without requiring any external base learner. By virtue of its inherent simplicity and lower computational cost, this embedded approach can be suitable for its adoption in online learning scenarios with severe resource constraints. Experiments with synthetic datasets show that the proposed technique is very competitive when compared to other drift detection techniques.
Effective air pollution prediction by combining time series decomposition with stacking and bagging ensembles of evolving spiking neural networks
(2023-12) Maciąg, Piotr S.; Bembenik, Robert; Piekarzewicz, Aleksandra; Del Ser, Javier; Lobo, Jesus L.; Kasabov, Nikola K.; IA
In this article, we introduce a new approach to air pollution prediction using the CEEMDAN time series decomposition method combined with the two-layered ensemble of predictors created based on the stacking and bagging techniques. The proposed ensemble approach is outperforming other selected state-of-the-art models when the bagging ensemble consisting of evolving Spiking Neural Networks (eSNNs) is used in the second layer of the stacking ensemble. In our experiments, we used the PM10 air pollution and weather dataset for Warsaw. As the results of the experiments show, the proposed ensemble can achieve the following error and agreement values over the tested dataset: error RMSE 6.91, MAE 5.14 and MAPE 21%; agreement IA 0.94. In addition, this article provides the computational and space complexity analysis of eSNNs predictors and offers a new encoding method for spiking neural networks that can be effectively applied for values of skewed distributions.
An Evolutionary Computation-Based Platform for Optimizing Infrastructure-as-Code Deployment Configurations
(Springer Science and Business Media Deutschland GmbH, 2024) Osaba, Eneko; Diaz-de-Arcaya, Josu; Alonso, Juncal; Lobo, Jesus L.; Benguria, Gorka; Etxaniz, Iñaki; Yang, Xin-She; Sherratt, R. Simon; Dey, Nilanjan; Joshi, Amit; Quantum; HPA; IA
PIACERE is an H2020 European project which objective is to implement a solution involving the development, deployment, and operation of Infrastructure-as-Code of applications running on cloud continuum. This technical paper is focused on describing a specific module of the whole PIACERE ecosystem: the IaC Optimizer Platform. The main objective of this component is to provide the user with optimized Infrastructure-as-Code configurations deployed on the most appropriate infrastructural elements that best meet the predefined requirements. For properly dealing with this problem, the IaC Optimizer Platform is based on Evolutionary Computation metaheuristics. More specifically, it resorts to NSGA-II and NSGA-III algorithms, depending on user needs. Additionally, we not only describe the IaC Optimizer Platform component in this paper, but we also show how it helps the user to find the most adequate Infrastructure-as-Code configurations.
Evolutionary Multi-Objective Quantization of Randomization-Based Neural Networks
(Institute of Electrical and Electronics Engineers Inc., 2023) Del Ser, Javier; Andres, Alain; Bilbao, Miren Nekane; Lana, Ibai; Lobo, Jesus L.; IA
The deployment of Machine Learning models on hardware devices has motivated a notable research activity around different strategies to alleviate their complexity and size. This is the case of neural architecture search or pruning in Deep Learning. This work places its focus on simplifying randomization-based neural networks by discovering fixed-point quantization policies that optimally balance the trade-off between performance and complexity reduction featured by these models. Specifically, we propose a combinatorial formulation of this problem, which we show to be efficiently solvable by multi-objective evolutionary algorithms. A benchmark for time series forecasting with Echo State Networks over 400 datasets reveals that high compression ratios can be achieved at practically admissible levels of performance degradation, showcasing the utility of the proposed problem formulation to deploy reservoir computing models on resource-constrained hardware devices.
Evolving Spiking Neural Networks for online learning over drifting data streams
(2018-12) Lobo, Jesus L.; Laña, Ibai; Del Ser, Javier; Bilbao, Miren Nekane; Kasabov, Nikola; IA
Nowadays huge volumes of data are produced in the form of fast streams, which are further affected by non-stationary phenomena. The resulting lack of stationarity in the distribution of the produced data calls for efficient and scalable algorithms for online analysis capable of adapting to such changes (concept drift). The online learning field has lately turned its focus on this challenging scenario, by designing incremental learning algorithms that avoid becoming obsolete after a concept drift occurs. Despite the noted activity in the literature, a need for new efficient and scalable algorithms that adapt to the drift still prevails as a research topic deserving further effort. Surprisingly, Spiking Neural Networks, one of the major exponents of the third generation of artificial neural networks, have not been thoroughly studied as an online learning approach, even though they are naturally suited to easily and quickly adapting to changing environments. This work covers this research gap by adapting Spiking Neural Networks to meet the processing requirements that online learning scenarios impose. In particular the work focuses on limiting the size of the neuron repository and making the most of this limited size by resorting to data reduction techniques. Experiments with synthetic and real data sets are discussed, leading to the empirically validated assertion that, by virtue of a tailored exploitation of the neuron repository, Spiking Neural Networks adapt better to drifts, obtaining higher accuracy scores than naive versions of Spiking Neural Networks for online learning environments.
Exploiting the stimuli encoding scheme of evolving Spiking Neural Networks for stream learning
(2020-03) Lobo, Jesus L.; Oregi, Izaskun; Bifet, Albert; Del Ser, Javier; IA; Quantum
Stream data processing has lately gained momentum with the arrival of new Big Data scenarios and applications dealing with continuously produced information flows. Unfortunately, traditional machine learning algorithms are not prepared to tackle the specific challenges imposed by data stream processing, such as the need for learning incrementally, limited memory and processing time requirements, and adaptation to non-stationary data, among others. To face these paradigms, Spiking Neural Networks have emerged as one of the most promising stream learning techniques, with variants such as Evolving Spiking Neural Networks capable of efficiently addressing many of these challenges. Interestingly, these networks resort to a particular population encoding scheme – Gaussian Receptive Fields – to transform the incoming stimuli into temporal spikes. The study presented in this manuscript sheds light on the predictive potential of this encoding scheme, focusing on how it can be applied as a computationally lightweight, model-agnostic preprocessing step for data stream learning. We provide informed intuition to unveil under which circumstances the aforementioned population encoding method yields effective prediction gains in data stream classification with respect to the case where no preprocessing is performed. Results obtained for a variety of stream learning models and both synthetic and real stream datasets are discussed to empirically buttress the capability of Gaussian Receptive Fields to boost the predictive performance of stream learning methods, spanning further research towards extrapolating our findings to other machine learning problems.
Identifying recommendation opportunities for computer-supported collaborative environments
(2016-10-01) Lobo, Jesus L.; Santos, Olga C.; Boticario, Jesus G.; Del Ser, Javier; IA
Collaborative indicators derived from participants' interactions can be used to support and improve their collaborative behaviour. In this research, we focus on automatically identifying recommendation opportunities in the Collaborative Logical Framework from participants' interactions. Different information sources have been considered: (a) statistical collaborative indicators; (b) social interactions; (c) opinions received by the participants via ratings; and (d) users' affective state and personality. The recommendations have been elicited considering the generality and transferability of the participants' interactions provided by the Collaborative Logical Framework. As a result, three scenarios have been identified that lead us to propose meaningful grouping suggestions and recommendations, which ultimately aimed to ground an informed personalized support to the participants in intensive collaborative frameworks.
IEM: A Unified Lifecycle Orchestrator for Multilingual IaC Deployments
(Association for Computing Machinery, Inc, 2023-04-15) Diaz-De-Arcaya, Josu; Osaba, Eneko; Benguria, Gorka; Etxaniz, Iñaki; Lobo, Jesus L.; Alonso, Juncal; Torre-Bastida, Ana I.; Almeida, Aitor; HPA; Quantum; IA
Over the last few years, DevOps methodologies have promoted a more streamlined operationalization of software components in production environments. Infrastructure as Code (IaC) technologies play a key role in the lifecycle management of applications, as they promote the delivery of the infrastructural elements alongside the application components. This way, IaC technologies aspire to minimize the problems associated with the environment by providing a repeatable and traceable process. However, there are a large variety of IaC frameworks, each of them focusing on a different phase of the operationalization lifecycle, hence the necessity to master numerous technologies. In this research, we present the IaC Execution Manager (IEM), a tool devoted to providing a unified framework for the operationalization of software components that encompasses the various stages and technologies involved in the application lifecycle. We analyze an industrial use case to improve the current approach and conclude the IEM is a suitable tool for solving the problem as it promotes automation, while reducing the learning curve associated with the required IaC technologies.
Lightweight Alternatives for Hyper-parameter Tuning in Drifting Data Streams
(IEEE Computer Society, 2021) Lobo, Jesus L.; Del Ser, Javier; Osaba, Eneko; Xue, Bing; Pechenizkiy, Mykola; Koh, Yun Sing; IA; Quantum
Scenarios dealing with data streams often undergo changes in data distribution, which ultimately lead to a performance degradation of algorithms learning from such data flows (concept drift). This phenomenon calls for the adoption of adaptive learning strategies for algorithms to perform resiliently after a change occurs. A multiplicity of approaches have so far addressed this issue by assorted means, e.g. instances weighting, ensembling, instance selection, or parameter tuning, among others. This latter strategy is often neglected as it requires a hyper-parameter tuning process that stream learning scenarios cannot computationally afford in most practical settings. Processing times and memory space are usually severely constrained, thus making the tuning phase unfeasible. Consequently, the research community has largely opted for other adaptive strategies with lower computational demands. This work outlines a new perspective to alleviate the hyper-parameter tuning process in the context of concept drift adaptation. We propose two simple and lightweight mechanisms capable of discovering competitive configurations of learning algorithms used for data stream classification. We compare its performance to that of a modern hyper-parametric search method (Successive Halving) over extensive experiments with synthetic and real datasets. We conclude that our proposed methods perform competitively, while consuming less processing time and memory.
Managing the unknown in machine learning: Definitions, related areas, recent advances, and prospects: Definitions, related areas, recent advances, and prospects
(2024-09-28) Barcina-Blanco, Marcos; Lobo, Jesus L.; Garcia-Bringas, Pablo; Del Ser, Javier; IA
In the rapidly evolving domain of machine learning, the ability to adapt to unforeseen circumstances and novel data types is of paramount importance. The deployment of Artificial Intelligence is progressively aimed at more realistic and open scenarios where data, tasks, and conditions are variable and not fully predetermined, and therefore where a closed set assumption cannot be hold. In such evolving environments, machine learning is asked to be autonomous, continuous, and adaptive, requiring effective management of uncertainty and the unknown to fulfill expectations. In response, there is a vigorous effort to develop a new generation of models, which are characterized by enhanced autonomy and a broad capacity to generalize, enabling them to perform effectively across a wide range of tasks. The field of machine learning in open set environments poses many challenges and also brings together different paradigms, some traditional but others emerging, where the overlapping and confusion between them makes it difficult to distinguish them or give them the necessary relevance. This work delves into the frontiers of methodologies that thrive in these open set environments, by identifying common practices, limitations, and connections between the paradigms Open-Ended Learning, Open-World Learning, Open Set Recognition, and other related areas such as Continual Learning, Out-of-Distribution detection, Novelty Detection, and Active Learning. We seek to easy the understanding of these fields and their common roots, uncover open problems and suggest several research directions that may motivate and articulate future efforts towards more robust and autonomous systems.
MO-MFCGA: Multiobjective Multifactorial Cellular Genetic Algorithm for Evolutionary Multitasking
(Institute of Electrical and Electronics Engineers Inc., 2021) Osaba, Eneko; Del Ser, Javier; Martinez, Aritz D.; Lobo, Jesus L.; Nebro, Antonio J.; Yang, Xin She; Quantum; IA
Multiobjetive optimization has gained a considerable momentum in the evolutionary computation scientific community. Methods coming from evolutionary computation have shown a remarkable performance for solving this kind of optimization problems thanks to their implicit parallelism and the simultaneous convergence towards the Pareto front. In any case, the resolution of multiobjective optimization problems (MOPs) from the perspective of multitasking optimization remains almost unexplored. Multitasking is an incipient research stream which explores how multiple optimization problems can be simultaneously addressed by performing a single search process. The main motivation behind this solving paradigm is to exploit the synergies between the different problems (or tasks) being optimized. Going deeper, we resort in this paper to the also recent paradigm Evolutionary Multitasking (EM). We introduce the adaptation of the recently proposed Multifactorial Cellular Genetic Algorithm (MFCGA) for solving MOPs, giving rise to the Multiobjective MFCGA (MO-MFCGA). An extensive performance analysis is conducted using the Multiobjective Multifactorial Evolutionary Algorithm as comparison baseline. The experimentation is conducted over 10 multitasking setups, using the Multiobjective Euclidean Traveling Salesman Problem as benchmarking problem. We also perform a deep analysis on the genetic transferability among the problem instances employed, using the synergies among tasks aroused along the MO-MFCGA search procedure.
Modelling gene interaction networks from time-series gene expression data using evolving spiking neural networks
(2020-12-01) Capecci, Elisa; Lobo, Jesus L.; Laña, Ibai; Espinosa-Ramos, Josafath I.; Kasabov, Nikola; IA
The genetic mechanisms responsible for the differentiation, metabolism, morphology and function of a cell in both normal and abnormal conditions can be uncovered by the analysis of transcriptomes. Mining big data such as the information encoded in nucleic acids, proteins, and metabolites has challenged researchers for several years now. Even though bioinformatics and system biology techniques have improved greatly and many improvements have been done in these fields of research, most of the processes that influence gene interaction over time are still unknown. In this study, we apply state-of-the art spiking neural network techniques to model, analyse and extract information about the regulatory processes of gene expression over time. A case study of microarray profiling in human skin during elicitation of eczema is used to examine the temporal association of genes involved in the inflammatory response, by means of a gene interaction network. Spiking neural network techniques are able to learn the interaction between genes using information encoded from the time-series gene expression data as spikes. The temporal interaction is learned, and the patterns of activity extracted and analysed with a gene interaction network. Results demonstrated that useful knowledge can be extracted from the data by using spiking neural network, unlocking some of the possible mechanisms involved in the regulatory process of gene expression.
Multi-objective heuristics applied to robot task planning for inspection plants
(Institute of Electrical and Electronics Engineers Inc., 2017-07-05) Landa-Torres, Itziar; Lobo, Jesus L.; Murua, Idoia; Manjarres, Diana; Del Ser, Javier; Tecnalia Research & Innovation; IA; HPA
Robotics are generally subject to stringent operational conditions that impose a high degree of criticality on the allocation of resources and the schedule of operations in mission planning. In this regard the so-called cost of a mission must be considered as an additional criterion when designing optimal task schedules within the mission at hand. Such a cost can be conceived as the impact of the mission on the robotic resources themselves, which range from the consumption of battery to other negative effects such as mechanic erosion. This manuscript focuses on this issue by presenting experimental results obtained over realistic scenarios of two heuristic solvers (MOHS and NSGA-II) aimed at efficiently scheduling tasks in robotic swarms that collaborate together to accomplish a mission. The heuristic techniques resort to a Random-Keys encoding strategy to represent the allocation of robots to tasks whereas the relative execution order of such tasks within the schedule of certain robots is computed based on the Traveling Salesman Problem (TSP). Experimental results in three different deployment scenarios reveal the goodness of the proposed technique based on the Multi-objective Harmony Search algorithm (MOHS) in terms of Hypervolume (HV) and Coverage Rate (CR) performance indicators.

Browsing by Author "Lobo, Jesus L."

Results Per Page

Sort Options