DI - GCERN - Artículos de revistas científicas

Permanent URI for this collection


Recent Submissions

Now showing 1 - 20 of 164
  • Publication
    AWS PredSpot: Machine Learning for Predicting the Price of Spot Instances in AWS Cloud
    (UNIR La Universidad en Internet, 2022-01-01) Baldominos Gomez, Alejandro; Baldominos Gómez, Alejandro; Sáez Achaerandio, Yago; Quintana, David; Isasi, Pedro; Comunidad de Madrid; Universidad Carlos III de Madrid
    Elastic Cloud Compute (EC2) is one of the most well-known services provided by Amazon for provisioning cloud computing resources, also known as instances. Besides the classical on-demand scheme, where users purchase compute capacity at a fixed cost, EC2 supports so-called spot instances, which are offered following a bidding scheme, where users can save up to 90% of the cost of the on-demand instance. EC2 spot instances can be a useful alternative for attaining an important reduction in infrastructure cost, but designing bidding policies can be a difficult task, since bidding under their cost will either prevent users from provisioning instances or losing those that they already own. Towards this extent, accurate forecasting of spot instance prices can be of an outstanding interest for designing working bidding policies. In this paper, we propose the use of different machine learning techniques to estimate the future price of EC2 spot instances. These include linear, ridge and lasso regressions, multilayer perceptrons, K-nearest neighbors, extra trees and random forests. The obtained performance varies significantly between instances types, and root mean squared errors ranges between values very close to zero up to values over 60 in some of the most expensive instances. Still, we can see that for most of the instances, forecasting performance is remarkably good, encouraging further research in this field of study
  • Publication
    Combinatorial versus sequential auctions to allocate PPP highway projects
    (Elservier, 2022-03-01) Mochon, Pablo; Mochon Saez, Maria Asuncion; Sáez Achaerandio, Yago
    This article models a procurement process for allocating multiple related public-private partnership (PPP) highway projects. Traditionally, public infrastructure procurement processes have used a sequential allocation mechanism, despite the potential benefits of allocating all projects at once. The main contribution of this research is to address the question whether these projects should be auctioned individually, in sequential auctions, or at the same time, in a combinatorial auction. Our goal is to understand the impact of the allocation process in terms of efficiency and social welfare. In sequential auctions, bidders submit their offers for each project independently. However, in combinatorial auctions, contractors have the ability to bid for their preferred packages (combinations of projects), reflecting synergies or entry costs, if any, in their valuations. We have compared the impact in terms of efficient allocation and social welfare of both mechanisms in order to help policymakers to take future decisions when facing these processes. The methodology used to address these core questions in the multidisciplinary environment described is based on social simulations, which involves conducting analysis by means of computational simulations. For this work we have created a sophisticated valuation model adapted to the public infrastructure sector and we have developed a simulator which includes multiple types of bidders, projects and several scenarios. The experimental setup is based on the second wave of the Colombian 4G program, a case involving the allocation of 9 highway construction projects across the country. We have also included references to multiple examples of real markets in which these mechanisms could be implemented. Therefore, this research provides a valuable reference for policymakers chasing to enhance market design that could be applied in many real-world scenarios. The results reveal that the combinatorial mechanism improves the process in terms of optimal allocation and efficiency, yielding significant savings for all parties.
  • Publication
    Pareto optimal prediction intervals with hypernetworks
    (Elsevier, 2023-01) Alcántara Mata, Antonio; Galván, Inés M.; Aler, Ricardo; Ministerio de Ciencia e Innovación (España); Agencia Estatal de Investigación (España)
    As the relevance of probabilistic forecasting grows, the need of estimating multiple high-quality prediction intervals (PI) also increases. In the current state of the art, most deep neural network gradient descent-based methods take into account interval width and coverage into a single loss function, focusing on a unique nominal coverage target, and adding additional parameters to control the coverage-width trade-off. The Pareto Optimal Prediction Interval Hypernetwork (POPI-HN) approach developed in this work has been derived to treat this coverage-width trade-off as a multi-objective problem, obtaining a complete set of Pareto Optimal solutions (Pareto front). POPI-HN are able to be trained through gradient descent with no need to add extra parameters to control the width-coverage trade-off of PIs. Once the Pareto set has been obtained, users can extract the PI with the required coverage. Comparative results with recently introduced Quality-Driven loss show similar behavior in coverage while improving interval width for the majority of the studied domains, making POPI-HN a competing alternative for estimating uncertainty in regression tasks where PIs with multiple coverages are needed.
  • Publication
    A survey of handwritten character recognition with MNIST and EMNIST
    (MDPI, 2019-08-01) Baldominos Gómez, Alejandro; Sáez Achaerandio, Yago; Isasi, Pedro; Ministerio de Educación, Cultura y Deporte (España)
    This paper summarizes the top state-of-the-art contributions reported on the MNIST dataset for handwritten digit recognition. This dataset has been extensively used to validate novel techniques in computer vision, and in recent years, many authors have explored the performance of convolutional neural networks (CNNs) and other deep learning techniques over this dataset. To the best of our knowledge, this paper is the first exhaustive and updated review of this dataset; there are some online rankings, but they are outdated, and most published papers survey only closely related works, omitting most of the literature. This paper makes a distinction between those works using some kind of data augmentation and works using the original dataset out-of-the-box. Also, works using CNNs are reported separately; as they are becoming the state-of-the-art approach for solving this problem. Nowadays, a significant amount of works have attained a test error rate smaller than 1% on this dataset; which is becoming non-challenging. By mid-2017, a new dataset was introduced: EMNIST, which involves both digits and letters, with a larger amount of data acquired from a database different than MNIST's. In this paper, EMNIST is explained and some results are surveyed.
  • Publication
    Coin.AI: A proof-of-useful-work scheme for blockchain-based distributed deep learning
    (MDPI, 2019-08) Baldominos Gómez, Alejandro; Sáez Achaerandio, Yago
    One decade ago, Bitcoin was introduced, becoming the first cryptocurrency and establishing the concept of "blockchain" as a distributed ledger. As of today, there are many different implementations of cryptocurrencies working over a blockchain, with different approaches and philosophies. However, many of them share one common feature: they require proof-of-work to support the generation of blocks (mining) and, eventually, the generation of money. This proof-of-work scheme often consists in the resolution of a cryptography problem, most commonly breaking a hash value, which can only be achieved through brute-force. The main drawback of proof-of-work is that it requires ridiculously large amounts of energy which do not have any useful outcome beyond supporting the currency. In this paper, we present a theoretical proposal that introduces a proof-of-useful-work scheme to support a cryptocurrency running over a blockchain, which we named Coin.AI. In this system, the mining scheme requires training deep learning models, and a block is only mined when the performance of such model exceeds a threshold. The distributed system allows for nodes to verify the models delivered by miners in an easy way (certainly much more efficiently than the mining process itself), determining when a block is to be generated. Additionally, this paper presents a proof-of-storage scheme for rewarding users that provide storage for the deep learning models, as well as a theoretical dissertation on how the mechanics of the system could be articulated with the ultimate goal of democratizing access to artificial intelligence.
  • Publication
    Using smart persistence and random forests to predict photovoltaic energy production
    (MDPI, 2019-01-01) Huertas Tato, Javier; Centeno Brito, Miguel; Ministerio de Ciencia e Innovación (España)
    Solar energy forecasting is an active research problem and a key issue to increase the competitiveness of solar power plants in the energy market. However, using meteorological, production, or irradiance data from the past is not enough to produce accurate forecasts. This article aims to integrate a prediction algorithm (Smart Persistence), irradiance, and past production data, using a state-of-the-art machine learning technique (Random Forests). Three years of data from six solar PV modules at Faro (Portugal) are analyzed. A set of features that combines past data, predictions, averages, and variances is proposed for training and validation. The experimental results show that using Smart Persistence as a Machine Learning input greatly improves the accuracy of short-term forecasts, achieving an NRMSE of 0.25 on the best panels at short horizons and 0.33 on a 6 h horizon.
  • Publication
    Editor's Note. Towards an intelligent society: advances in marketing and neuroscience
    (UNIR, 2019-09) Mochón Morcillo, Francisco; Baldominos Gómez, Alejandro
    This Special Issue focuses in cases that explore the relationship between Artificial Intelligence and marketing, as well as neuroscience. AI can be combined with specific neuroscience techniques to achieve a more successful and profitable neuromarketing. For this Special Issue, we have found that descriptions of successful use cases are highly valuable to help researchers identify fields where novel applications of AI can enhance the outcome of digital marketing and neuroscience.
  • Publication
    Hybridizing evolutionary computation and deep neural networks: an approach to handwriting recognition using committees and transfer learning
    (Hindawi, 2019-03-26) Baldominos Gómez, Alejandro; Sáez Achaerandio, Yago; Isasi, Pedro; Ministerio de Educación, Cultura y Deporte (España)
    Neuroevolution is the field of study that uses evolutionary computation in order to optimize certain aspect of the design of neural networks, most often its topology and hyperparameters. The field was introduced in the late-1980s, but only in the latest years the field has become mature enough to enable the optimization of deep learning models, such as convolutional neural networks. In this paper, we rely on previous work to apply neuroevolution in order to optimize the topology of deep neural networks that can be used to solve the problem of handwritten character recognition. Moreover, we take advantage of the fact that evolutionary algorithms optimize a population of candidate solutions, by combining a set of the best evolved models resulting in a committee of convolutional neural networks. This process is enhanced by using specific mechanisms to preserve the diversity of the population. Additionally, in this paper, we address one of the disadvantages of neuroevolution: the process is very expensive in terms of computational time. To lessen this issue, we explore the performance of topology transfer learning: whether the best topology obtained using neuroevolution for a certain domain can be successfully applied to a different domain. By doing so, the expensive process of neuroevolution can be reused to tackle different problems, turning it into a more appealing approach for optimizing the design of neural networks topologies. After evaluating our proposal, results show that both the use of neuroevolved committees and the application of topology transfer learning are successful: committees of convolutional neural networks are able to improve classification results when compared to single models, and topologies learned for one problem can be reused for a different problem and data with a good performance. Additionally, both approaches can be combined by building committees of transferred topologies, and this combination attains results that combine the best of both
  • Publication
    Digital teaching materials and their relationship with the metacognitive skills of students in primary education
    (MDPI, 2020-01-01) Nieto Marquez, Natalia Lara; Baldominos Gómez, Alejandro; Perez Nieto, Miguel Angel; Comunidad de Madrid
    Metacognition is a construct that is noteworthy for its relationship with the prediction and enhancement of student performance. It is of interest in education, as well as in the field of cognitive psychology, because it contributes to competencies, such as learning to learn and the understanding of information. This study conducted research at a state school in the Community of Madrid (Spain) with a sample of 130 students in Grade 3 of their primary education (8 years old). The research involved the use of a digital teaching platform called Smile and Learn, as the feedback included in the digital activities may have an effect on students' metacognition. We analyzed the implementation of the intelligent platform at school and the activities most commonly engaged in. The Junior Metacognitive Awareness Inventory (Jr. MAI) was the measuring instrument chosen for the external evaluation of metacognition. The study's results show a higher use of logic and spatial activities. A relationship is observed between the use of digital exercises that have specific feedback and work on logic and visuospatial skills with metacognitive knowledge. We discuss our findings surrounding educational implications, metacognition assessment, and recommendations for improvements of the digital materials.
  • Publication
    Assessment of the effects of digital educational material on executive function performance
    (Frontiers, 2020-11-23) Nieto-Márquez, Natalia Lara; Cardeña Martinez, Alejandro; Baldominos Gómez, Alejandro; Gonzalez Petronila, Almudena; Pérez Nieto, Miguel Ángel; Comunidad de Madrid
    The Learning Analytics system of the Smile and Learn platform recorded the students'use during class. According to the usage analysis, the results obtained show preference of using activities from Logic and Spatial worlds. In the external analysis of the effect of the learning material, the results record a significant effect using activities in Logic and Spatial worlds with the Gray Trails task, which involves spatial perception, processing speed, and working memory, among others. A second analysis to contrast the results with a post hoc design approaches relationships among executive functions as involved in tasks like Gray Trails, Interference, and Ring Tasks within the usage of Spatial and Logic activities. The need for further research to improve these materials for enhanced learning and the extrapolation of training from executive functions to other tasks is discussed. Likewise, limitations of the implementation and design of these materials are pointed out.
  • Publication
    An exploratory analysis of the implementation and use of an intelligent platform for learning in primary education
    (MDPI, 2020-01-01) Nieto Marquez, Natalia Lara; Baldominos Gómez, Alejandro; Cardeña Martinez, Alejandro; Perez Nieto, Miguel Angel; Comunidad de Madrid
    Smile and Learn is an intelligent platform with more than 4500 educational activities for children aged 3-12. The digital material developed covers all courses of primary education and most of the subjects with the different topic-related worlds with activities in the field of logics and mathematics, science, linguistics and tales, visual-spatial and cognitive skills, emotional intelligence, arts, and multiplayer games. This kind of material supports active learning and new pedagogical models for teachers to use in their lessons. The purpose of this paper is to explore the usage of the platform in three pilot groups schools from different regions of Spain, outlining future directions in the design of such digital materials. Usage is assessed via descriptive analysis and variance analysis, with data collected from users interacting with the intelligent platform. The results show a high use of STEM (Science, Technology, Engineering and Maths) activities among all the activities that could be chosen. Cross-curricular activities are also used. Continuation in the development of such materials is concluded necessary, focusing integration of different fields, accentuating games over quizzes, and the value of teacher training for improving their use in schools.
  • Publication
    Improving Solar Radiation Nowcasts by Blending Data-Driven, Satellite-Images-Based and All-Sky-Imagers-Based Models Using Machine Learning Techniques
    (MDPI, 2023-04-28) López Cuesta, Miguel; Aler, Ricardo; Galván, Inés M.; Pozo Vázquez, David; Comunidad de Madrid; European Commission; Ministerio de Ciencia e Innovación (España)
    Accurate solar radiation nowcasting models are critical for the integration of the increasing solar energy in power systems. This work explored the benefits obtained by the blending of four all-sky-imagers (ASI)-based models, two satellite-images-based models and a data-driven model. Two blending approaches (general and horizon) and two blending models (linear and random forest (RF)) were evaluated. The relative contribution of the different forecasting models in the blendedmodels-derived benefits was also explored. The study was conducted in Southern Spain; blending models provide one-minute resolution 90 min-ahead GHI and DNI forecasts. The results show that the general approach and the RF blending model present higher performance and provide enhanced forecasts. The improvement in rRMSE values obtained by model blending was up to 30% for GHI (40% for DNI), depending on the forecasting horizon. The greatest improvement was found at lead times between 15 and 30 min, and was negligible beyond 50 min. The results also show that blending models using only the data-driven model and the two satellite-images-based models (one using high resolution images and the other using low resolution images) perform similarly to blending models that used the ASI-based forecasts. Therefore, it was concluded that suitable model blending might prevent the use of expensive (and highly demanding, in terms of maintenance) ASI-based systems for point nowcasting
  • Publication
    Electricity market integration and impact of renewable energy sources in the Central Western Europe region: Evolution since the implementation of the Flow-Based Market Coupling mechanism
    (Elsevier, 2022-11) Corona, Luis; Mochón, Asunción; Sáez Achaerandio, Yago
    The wholesale electricity markets in Europe are undergoing major changes. The pursuit of a major integration for the development of the Internal Energy market is the main driver of this transformation. In this context, the Central Western Europe region implemented for the first time in May 2015 a new and complex mechanism to manage cross-border capacity allocation in the frame of the day-ahead market: the Flow-Based Market Coupling. This paper is the first to consider recent data of the electricity market to develop a predictive model for identifying convergence and congestion situations in the Central Western Europe region. The occurrence of convergence or not is considered as a binary outcome to build a probit model. This model combines meaningful information about the electricity market features of the interconnected countries between 2016 and 2018, including for the first time data on renewable energy sources forecasts for this type of studies. The results of this study identify the major role that Germany and France play in the integration process. The estimated coefficients of the model show that the strong development of solar and wind power in Germany appears as an important driver of congestion in the Central Western Europe. In order to improve the benefits of market integration and the further development of the Internal Energy market, the results of this study encourage policy-makers to promote cooperation and coordination among all the actors involved in the electricity market and to pursue the expansion of the power grid to minimize the occurrence of congestion situations.
  • Publication
    Assessment of COVID-19's impact on EdTech: Case study on an educational platform, architecture and teachers' experience
    (MDPI, 2022-10) Nieto-Márquez, Natalia Lara; Baldominos Gómez, Alejandro; Iglesias Solían, Manuel; Martin Dobón, Elisa; Zuluaga Arevalo, J. Alexandra
    The education sector has been confronted with different challenges due to the situation caused by the pandemic, when families were asked to be confined at home as well as return when schools were opened. This lockdown situation presented both a challenge for the EdTech sector and for teachers and families. Consequently, this study analyzes the importance of online methodologies, usage of an educational resource example, and the impact of the lockdown. Thus, these objectives are assessed from different perspectives such as users' consumption, technical challenges of cloud architecture and experience from teachers who have used the platform during the lockdown. In this way, to understand the challenges of Cloud architecture, the structure of the Pre-COVID-19 platform and the changes implemented to adapt to the new needs are described. An increase in schools' subscriptions was observed when home lockdown was decreed, the differences in usage with the return to the classroom are also discussed. The research methodology entailed an assessment instrument developed for teachers. Teachers highlight the contents of Smile and Learn platform and their motivational characteristics for the students' learning. The assessment points out the limitations that many teachers face while using these resources.
  • Publication
    A combination of supervised dimensionality reduction and learning methods to forecast solar radiation
    (Springer, 2022-10-06) García Cuesta, Esteban; Aler, Ricardo; Pozo Vázquez, David; Galván, Inés M.; Comunidad de Madrid; Agencia Estatal de Investigación (España)
    Machine learning is routinely used to forecast solar radiation from inputs, which are forecasts of meteorological variables provided by numerical weather prediction (NWP) models, on a spatially distributed grid. However, the number of features resulting from these grids is usually large, especially if several vertical levels are included. Principal Components Analysis (PCA) is one of the simplest and most widely-used methods to extract features and reduce dimensionality in renewable energy forecasting, although this method has some limitations. First, it performs a global linear analysis, and second it is an unsupervised method. Locality Preserving Projection (LPP) overcomes the locality problem, and recently the Linear Optimal Low-Rank (LOL) method has extended Linear Discriminant Analysis (LDA) to be applicable when the number of features is larger than the number of samples. Supervised Nonnegative Matrix Factorization (SNMF) also achieves this goal extending the Nonnegative Matrix Factorization (NMF) framework to integrate the logistic regression loss function. In this article we try to overcome all these issues together by proposing a Supervised Local Maximum Variance Preserving (SLMVP) method, a supervised non-linear method for feature extraction and dimensionality reduction. PCA, LPP, LOL, SNMF and SLMVP have been compared on Global Horizontal Irradiance (GHI) and Direct Normal Irradiance (DNI) radiation data at two different Iberian locations: Seville and Lisbon. Results show that for both kinds of radiation (GHI and DNI) and the two locations, SLMVP produces smaller MAE errors than PCA, LPP, LOL, and SNMF, around 4.92% better for Seville and 3.12% for Lisbon. It has also been shown that, although SLMVP, PCA, and LPP benefit from using a non-linear regression method (Gradient Boosting in this work), this benefit is larger for PCA and LPP because SMLVP is able to perform non-linear transformations of inputs.
  • Publication
    Deep neural networks for the quantile estimation of regional renewable energy production
    (Springer, 2022-08-02) Alcántara Mata, Antonio; Galván, Inés M.; Aler, Ricardo; Comunidad de Madrid; Ministerio de Ciencia e Innovación (España); Agencia Estatal de Investigación (España)
    Wind and solar energy forecasting have become crucial for the inclusion of renewable energy in electrical power systems. Although most works have focused on point prediction, it is currently becoming important to also estimate the forecast uncertainty. With regard to forecasting methods, deep neural networks have shown good performance in many fields. However, the use of these networks for comparative studies of probabilistic forecasts of renewable energies, especially for regional forecasts, has not yet received much attention. The aim of this article is to study the performance of deep networks for estimating multiple conditional quantiles on regional renewable electricity production and compare them with widely used quantile regression methods such as the linear, support vector quantile regression, gradient boosting quantile regression, natural gradient boosting and quantile regression forest methods. A grid of numerical weather prediction variables covers the region of interest. These variables act as the predictors of the regional model. In addition to quantiles, prediction intervals are also constructed, and the models are evaluated using different metrics. These prediction intervals are further improved through an adapted conformalized quantile regression methodology. Overall, the results show that deep networks are the best performing method for both solar and wind energy regions, producing narrow prediction intervals with good coverage.
  • Publication
    Direct estimation of prediction intervals for solar and wind regional energy forecasting with deep neural networks
    (Elsevier, 2022-09-01) Alcántara Mata, Antonio; Galván, Inés M.; Aler, Ricardo; Comunidad de Madrid; Universidad Carlos III de Madrid
    Deep neural networks (DNN) are becoming increasingly relevant for probabilistic forecasting because of their ability to estimate prediction intervals (PIs). Two different ways for estimating PIs with neural networks stand out: quantile estimation for posterior PI construction and direct PI estimation. The former first estimates quantiles, which are then used to construct PIs, while the latter directly obtains the lower and upper PI bounds by optimizing some loss functions, with the advantage that PI width is directly considered in the optimization process and thus may result in narrower intervals. In this work, two different DNN-based models are studied for direct PI estimation, and compared with DNN for quantile estimation in the context of solar and wind regional energy forecasting. The first approach is based on the recent quality-driven loss and is formulated to estimate multiple PIs with a single model. The second is a novel approach that employs hypernetworks (HN), where direct PI estimation is formulated as a multi-objective problem, returning a Pareto front of solutions that contains all possible coverage-width optimal trade-offs. This formulation allows HN to obtain optimal PIs for all possible coverages without increasing the number of network outputs or adjusting additional hyperparameters, as opposed to the first direct model. Results show that prediction intervals from direct estimation are narrower (up to 20%) than those of quantile estimation, for target coverages 70%–80% for all regions, and also 85%, 90%, and 95% depending on the region, while HN always achieves the required coverage for the higher target coverages.
  • Publication
    Incremental market behavior classification in presence of recurring concepts
    (MDPI, 2019-01-01) Suárez Cetrulo, Andrés L.; Cervantes Rovira, Alejandro; Quintana, David; Ministerio de Economía y Competitividad (España)
    In recent years, the problem of concept drift has gained importance in the financial domain. The succession of manias, panics and crashes have stressed the non-stationary nature and the likelihood of drastic structural or concept changes in the markets. Traditional systems are unable or slow to adapt to these changes. Ensemble-based systems are widely known for their good results predicting both cyclic and non-stationary data such as stock prices. In this work, we propose RCARF (Recurring Concepts Adaptive Random Forests), an ensemble tree-based online classifier that handles recurring concepts explicitly. The algorithm extends the capabilities of a version of Random Forest for evolving data streams, adding on top a mechanism to store and handle a shared collection of inactive trees, called concept history, which holds memories of the way market operators reacted in similar circumstances. This works in conjunction with a decision strategy that reacts to drift by replacing active trees with the best available alternative: either a previously stored tree from the concept history or a newly trained background tree. Both mechanisms are designed to provide fast reaction times and are thus applicable to high-frequency data. The experimental validation of the algorithm is based on the prediction of price movement directions one second ahead in the SPDR (Standard & Poor's Depositary Receipts) S&P 500 Exchange-Traded Fund. RCARF is benchmarked against other popular methods from the incremental online machine learning literature and is able to achieve competitive results.
  • Publication
    Dynamic generation of investment recommendations using grammatical evolution
    (Universidad Internacional de la Rioja (UNIR), 2021-04-22) Martín Fernández, Carlos; Quintana, David; Isasi, Pedro; Comunidad de Madrid; Universidad Carlos III de Madrid
    The attainment of trading rules using Grammatical Evolution traditionally follows a static approach. A single rule is obtained and then used to generate investment recommendations over time. The main disadvantage of this approach is that it does not consider the need to adapt to the structural changes that are often associated with financial time series. We improve the canonical approach introducing an alternative that involves a dynamic selection mechanism that switches between an active rule and a candidate one optimized for the most recent market data available. The proposed solution seeks the flexibility required by structural changes while limiting the transaction costs commonly associated with constant model updates. The performance of the algorithm is compared with four alternatives: the standard static approach; a sliding window-based generation of trading rules that are used for a single time period, and two ensemble-based strategies. The experimental results, based on market data, show that the suggested approach beats the rest.
  • Publication
    Convolution on neural networks for high-frequency trend prediction of cryptocurrency exchange rates using technical indicators
    (Elsevier, 2020-07-01) Alonso Monsalve, Saúl; Suárez Cetrulo, Andrés L.; Cervantes Rovira, Alejandro; Quintana, David
    This study explores the suitability of neural networks with a convolutional component as an alternative to traditional multilayer perceptrons in the domain of trend classification of cryptocurrency exchange rates using technical analysis in high frequencies. The experimental work compares the performance of four different network architectures -convolutional neural network, hybrid CNN-LSTM network, multilayer perceptron and radial basis function neural network- to predict whether six popular cryptocurrencies -Bitcoin, Dash, Ether, Litecoin, Monero and Ripple- will increase their value vs. USD in the next minute. The results, based on 18 technical indicators derived from the exchange rates at a one-minute resolution over one year, suggest that all series were predictable to a certain extent using the technical indicators. Convolutional LSTM neural networks outperformed all the rest significantly, while CNN neural networks were also able to provide good results specially in the Bitcoin, Ether and Litecoin cryptocurrencies.