Search

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. We would also like to acknowledge the financial support of the Spanish Ministry of Science, Innovation and Universities under grant PGC2018-096849-B-I00 (MCFin)

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 &amp, Poor&rsquo, s Depositary Receipts) S&amp, 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.

We have compared the performance of different machine learning techniques for human activity recognition. Experiments were made using a benchmark dataset where each subject wore a device in the pocket and another on the wrist. The dataset comprises thirteen activities, including physical activities, common postures, working activities and leisure activities. We apply a methodology known as the activity recognition chain, a sequence of steps involving preprocessing, segmentation, feature extraction and classification for traditional machine learning methods, we also tested convolutional deep learning networks that operate on raw data instead of using computed features. Results show that combination of two sensors does not necessarily result in an improved accuracy. We have determined that best results are obtained by the extremely randomized trees approach, operating on precomputed features and on data obtained from the wrist sensor. Deep learning architectures did not produce competitive results with the tested architecture.

This paper tries to tackle the modern challenge of practical steganalysis over large data by presenting a novel approach whose aim is to perform with perfect accuracy and in a completely automatic manner. The objective is to detect changes introduced by the steganographic process in those data objects, including signatures related to the tools being used. Our approach achieves this by first extracting reliable regularities by analyzing pairs of modified and unmodified data objects; then, combines these findings by creating general patterns present on data used for training. Finally, we construct a Naive Bayes model that is used to perform classification, and operates on attributes extracted using the aforementioned patterns. This technique has been be applied for different steganographic tools that operate in media files of several types. We are able to replicate or improve on a number or previously published results, but more importantly, we in addition present new steganalytic findings over a number of popular tools that had no previous known attacks. This work was supported by Engineering and Physical Sciences Research Council (EPSRC) Grant EP/N024192/1 (J.C.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. This work was supported by European Union’s Horizon 2020 Research and Innovation Programme, Grant agreement No.700326.

The home care crew scheduling (HCCS) problem is a planning task whose goal is to allocate a set of professional caregivers in the most efficient way to perform a number of assistencial and health care visits to the customers private homes. This is part of an important trend in advanced health care systems, to promote "independent living" specially in situations of dependency on long-term care. This not only ensures a higher quality of life but also a lower cost for society. Real instances of the HCCS problem are large and highly constrained due to both caregivers' contract limitations and customers' needs. This paper presents an advanced parallel model that solves HCCS problems using a grid-based asynchronous evolutionary algorithm (EA). Our approach has been tested using a grid computing facility of up to 300 nodes. The algorithm is a modified (1 + lambda) EA, parallelized using a master/worker model that minimizes communication requirements and processor bottlenecks by distributing both the execution of the EA operators and the evaluation of solutions. We have used three large real-world instances provided by a private company to perform experimentation with different configurations of the EA and number of workers. Results show that our algorithm achieves solutions that clearly outperform the solution provided by the company and the grid-based algorithm is able to handle real world HCCS problems The work of Francisco Luna and Juan F. Valenzuela is funded by the Spanish Ministry of Economy, Industry and Competitiveness under contract TIN2016-75097-P. The work of Alejandro Cervantes and Pedro Isasi is funded by the Spanish Ministry of Science and Innovation under contract TIN2011-28336.

In the context of forecasting for renewable energy, it is common to produce point forecasts but it is also important to have information about the uncertainty of the forecast. To this extent, instead of providing a single measure for the prediction, lower and upper bound for the expected value for the solar radiation are used (prediction interval). This estimation of optimal prediction intervals requires the simultaneous optimization of two objective measures: on one hand, it is important that the coverage probability of the interval is as close as possible to a given target value. On the other, in order to bound uncertainty, intervals must be narrow; this means that there is a trade-off between both objectives, as narrow intervals reduce the coverage probability for those solutions, as the actual value of solar radiation is more likely to fall outside the predicted margins. In this work we propose a multi-objective evolutionary approach that is able to optimize both goals simultaneously. The proposal uses neural networks to create complex non-linear models whose outputs are the upper and lower limits of the prediction intervals. Results have been compared with a single-objective optimization of similar neural network architectures and a baseline algorithm (quantile estimation with gradient boosting). Results show that the neural network is able to provide accurate results. Also, the multi-objective approach is able to obtain the best results and has also the advantage that a single run is able to obtain prediction intervals for any target coverage probability. This work has been funded by the Spanish Ministry of Science under contract ENE2014-56126-C2-2-R (AOPRIN-SOL project).

Stream Processing has recently become one of the current commercial trends to face huge amounts of data. However, normally these techniques need specific infrastructures and high resources in terms of memory and computing nodes. This paper shows how mini-batch techniques and topology extraction methods can help making gigabytes of data to be manageable for just one server using computationally costly Machine Learning techniques as Support Vector Machines. The algorithm iGNGSVM is proposed to improve the performance of Support Vector Machines in datasets where the data is continuously arriving. It is benchmarked against a mini-batch version of LibSVM, achieving good accuracy rates and performing faster than this.

Proceeding of: 2013 IEEE Congress on Evolutionary Computation (CEC), Cancun, 20-23 June 2013 Learning from non-stationary data requires methods that are able to deal with a continuous stream of data instances, possibly of infinite size, where the class distributions are potentially drifting over time. For handling such datasets, we are proposing a new method that incrementally creates and adapts a network of prototypes for classifying complex data received in an online fashion. The algorithm includes both an accuracy-based and time-based forgetting mechanisms that ensure that the model size does not grow indefinitely with large datasets. We have performed tests on seven benchmarking datasets for comparing our proposal with several approaches found in the literature, including ensemble algorithms associated to two different base classifiers. Performances obtained show that our algorithm is comparable to the best of the ensemble classifiers in terms of accuracy/time trade-off. Moreover, our approach appears to have significant advantages for dealing with data that has a complex, non-linearly separable topology. Spanish Ministry of Science and Innovation under the project MOVES, grant reference TIN2011-28336, and NSERC-Canada This article has been funded by the Spanish Ministry of Science and Innovation under the project MOVES with grant reference TIN2011-28336, and NSERC-Canada. Publicado

Proceeding of: 5th International Conference, EMO 2009, Nantes, France, April 7-10, 2009 This work presents the application of a parallel coopera- tive optimization approach to the broadcast operation in mobile ad-hoc networks (manets). The optimization of the broadcast operation im- plies satisfying several objectives simultaneously, so a multi-objective approach has been designed. The optimization lies on searching the best configurations of the dfcn broadcast protocol for a given manet sce- nario. The cooperation of a team of multi-objective evolutionary al- gorithms has been performed with a novel optimization model. Such model is a hybrid parallel algorithm that combines a parallel island- based scheme with a hyperheuristic approach. Results achieved by the algorithms in different stages of the search process are analyzed in order to grant more computational resources to the most suitable algorithms. The obtained results for a manets scenario, representing a mall, demon- strate the validity of the new proposed approach. This work has been supported by the ec (feder) and the Spanish Ministry of Education and Science inside the ‘Plan Nacional de i+d+i’ (tin2005-08818-c04) and (tin2008-06491-c04-02). The work of Gara Miranda has been developed under grant fpu-ap2004-2290. Publicado

This paper presents a new approach to Particle Swarm Optimization, called Michigan Approach PSO (MPSO), and its applica- tion to continuous classi cation problems as a Nearest Prototype (NP) classi er. In Nearest Prototype classi ers, a collection of prototypes has to be found that accurately represents the input patterns. The classi er then assigns classes based on the nearest prototype in this collection. The MPSO algorithm is used to process training data to nd those prototypes. In the MPSO algorithm each particle in a swarm represents a single pro- totype in the solution and it uses modi ed movement rules with particle competition and cooperation that ensure particle diversity. The proposed method is tested both with arti cial problems and with real benchmark problems and compared with several algorithms of the same family. Re- sults show that the particles are able to recognize clusters, nd decision boundaries and reach stable situations that also retain adaptation po- tential. The MPSO algorithm is able to improve the accuracy of 1-NN classi ers, obtains results comparable to the best among other classi ers, and improves the accuracy reported in literature for one of the problems. Publicado

IEEE Swarm Intelligence Symposium. Honolulu, HI, 1-5 april 2007 This paper presents an application of particle swarm optimization (PSO) to continuous classification problems, using a Michigan approach. In this work, PSO is used to process training data to find a reduced set of prototypes to be used to classify the patterns, maintaining or increasing the accuracy of the nearest neighbor classifiers. The Michigan approach PSO represents each prototype by a particle and uses modified movement rules with particle competition and cooperation that ensure particle diversity. The result is that the particles are able to recognize clusters, find decision boundaries and achieve stable situations that also retain adaptation potential. The proposed method is tested both with artificial problems and with three real benchmark problems with quite promising results.