arXiv subjects

Javier Del Ser

Publications and source records attributed to Javier Del Ser.

102 records · Page 6Linked to original sources

arXiv2019

Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI

In the last years, Artificial Intelligence (AI) has achieved a notable momentum that may deliver the best of expectations over many application sectors across the field. For this to occur, the entire community stands in front of the barrier of explainability, an inherent problem of AI techniques brought by sub-symbolism (e.g. ensembles or Deep Neural Networks) that were not present in the last hype of AI. Paradigms underlying this problem fall within the so-called eXplainable AI (XAI) field, which is acknowledged as a crucial feature for the practical deployment of AI models. This overview examines the existing literature in the field of XAI, including a prospect toward what is yet to be reached. We summarize previous efforts to define explainability in Machine Learning, establishing a novel definition that covers prior conceptual propositions with a major focus on the audience for which explainability is sought. We then propose and discuss about a taxonomy of recent contributions related to the explainability of different Machine Learning models, including those aimed at Deep Learning methods for which a second taxonomy is built. This literature analysis serves as the background for a series of challenges faced by XAI, such as the crossroads between data fusion and explainability. Our prospects lead toward the concept of Responsible Artificial Intelligence, namely, a methodology for the large-scale implementation of AI methods in real organizations with fairness, model explainability and accountability at its core. Our ultimate goal is to provide newcomers to XAI with a reference material in order to stimulate future research advances, but also to encourage experts and professionals from other disciplines to embrace the benefits of AI in their activity sectors, without any prior bias for its lack of interpretability.

cs.AI↗

arXiv2019

The Internet of Things: a Survey and Outlook

The recent history has witnessed disruptive advances in disciplines related to information and communication technologies that have laid a rich technological ecosystem for the growth and maturity of latent paradigms in this domain. Among them, sensor networks have evolved from the originally conceived set-up where hundreds of nodes with sensing and actuating functionalities were deployed to capture information from their environment and act accordingly (coining the so-called wireless sensor network concept) to the provision of such functionalities embedded in quotidian objects that communicate and work together to collaboratively accomplish complex tasks based on the information they acquire by sensing the environment. This is nowadays a reality, embracing the original idea of an Internet of things (IoT) forged in the late twentieth century, yet featuring unprecedented scales, capabilities and applications ignited by new radio interfaces, communication protocols and intelligent data-based models. This chapter examines the latest findings reported in the literature around these topics, with a clear focus on IoT communications, protocols and platforms, towards ultimately identifying opportunities and trends that will be at the forefront of IoT-related research in the near future.

cs.NI↗

arXiv2019

Exploiting a Stimuli Encoding Scheme of Spiking Neural Networks for Stream Learning

Stream data processing has gained progressive momentum with the arriving of new stream applications and big data scenarios. One of the most promising techniques in stream learning is the Spiking Neural Network, and some of them use an interesting population encoding scheme to transform the incoming stimuli into spikes. This study sheds lights on the key issue of this encoding scheme, the Gaussian receptive fields, and focuses on applying them as a pre-processing technique to any dataset in order to gain representativeness, and to boost the predictive performance of the stream learning methods. Experiments with synthetic and real data sets are presented, and lead to confirm that our approach can be applied successfully as a general pre-processing technique in many real cases.

cs.NE↗

arXiv2019

Spiking Neural Networks and Online Learning: An Overview and Perspectives

Applications that generate huge amounts of data in the form of fast streams are becoming increasingly prevalent, being therefore necessary to learn in an online manner. These conditions usually impose memory and processing time restrictions, and they often turn into evolving environments where a change may affect the input data distribution. Such a change causes that predictive models trained over these stream data become obsolete and do not adapt suitably to new distributions. Specially in these non-stationary scenarios, there is a pressing need for new algorithms that adapt to these changes as fast as possible, while maintaining good performance scores. Unfortunately, most off-the-shelf classification models need to be retrained if they are used in changing environments, and fail to scale properly. Spiking Neural Networks have revealed themselves as one of the most successful approaches to model the behavior and learning potential of the brain, and exploit them to undertake practical online learning tasks. Besides, some specific flavors of Spiking Neural Networks can overcome the necessity of retraining after a drift occurs. This work intends to merge both fields by serving as a comprehensive overview, motivating further developments that embrace Spiking Neural Networks for online learning scenarios, and being a friendly entry point for non-experts.

cs.NE↗

arXiv2019

Real-time Electrical Power Prediction in a Combined Cycle Power Plant

The prediction of electrical power in combined cycle power plants is a key challenge in the electrical power and energy systems field. This power output can vary depending on environmental variables, such as temperature, pressure, and humidity. Thus, the business problem is how to predict the power output as a function of these environmental conditions in order to maximize the profit. The research community has solved this problem by applying machine learning techniques and has managed to reduce the computational and time costs in comparison with the traditional thermodynamical analysis. Until now, this challenge has been tackled from a batch learning perspective in which data is assumed to be at rest, and where models do not continuously integrate new information into already constructed models. We present an approach closer to the Big Data and Internet of Things paradigms in which data is arriving continuously and where models learn incrementally, achieving significant enhancements in terms of data processing (time, memory and computational costs), and obtaining competitive performances. This work compares and examines the hourly electrical power prediction of several streaming regressors, and discusses about the best technique in terms of time processing and performance to be applied on this streaming scenario.

eess.SP↗

arXiv2019

Mobile XR over 5G: A way forward with mmWaves and Edge

This e-letter summarizes our most recent work and contributed approaches to increase the capacity, cut down on the latency and provide higher reliability in several extended reality (XR) scenarios. To that end, several technologies from emerging 5G communications systems are weaved together towards enabling a fully immersive XR experience.

eess.SP↗

arXiv2019

jMetalPy: a Python Framework for Multi-Objective Optimization with Metaheuristics

This paper describes jMetalPy, an object-oriented Python-based framework for multi-objective optimization with metaheuristic techniques. Building upon our experiences with the well-known jMetal framework, we have developed a new multi-objective optimization software platform aiming not only at replicating the former one in a different programming language, but also at taking advantage of the full feature set of Python, including its facilities for fast prototyping and the large amount of available libraries for data processing, data analysis, data visualization, and high-performance computing. As a result, jMetalPy provides an environment for solving multi-objective optimization problems focused not only on traditional metaheuristics, but also on techniques supporting preference articulation and dynamic problems, along with a rich set of features related to the automatic generation of statistical data from the results generated, as well as the real-time and interactive visualization of the Pareto front approximations produced by the algorithms. jMetalPy offers additionally support for parallel computing in multicore and cluster systems. We include some use cases to explore the main features of jMetalPy and to illustrate how to work with it.

cs.NE↗

arXiv2017

Millimeter Wave V2V Communications: Distributed Association and Beam Alignment

Recently millimeter-wave bands have been postulated as a means to accommodate the foreseen extreme bandwidth demands in vehicular communications, which result from the dissemination of sensory data to nearby vehicles for enhanced environmental awareness and improved safety level. However, the literature is particularly scarce in regards to principled resource allocation schemes that deal with the challenging radio conditions posed by the high mobility of vehicular scenarios. In this work we propose a novel framework that blends together Matching Theory and Swarm Intelligence to dynamically and efficiently pair vehicles and optimize both transmission and reception beamwidths. This is done by jointly considering Channel State Information (CSI) and Queue State Information (QSI) when establishing vehicle-to-vehicle (V2V) links. To validate the proposed framework, simulation results are presented and discussed where the throughput performance as well as the latency/reliability trade-offs of the proposed approach are assessed and compared to several baseline approaches recently proposed in the literature. The results obtained in our study show performance gains in terms of reliability and delay up to 25% for ultra-dense vehicular scenarios and on average 50% more paired vehicles that some of the baselines. These results shed light on the operational limits and practical feasibility of mmWave bands, as a viable radio access solution for future high-rate V2V communications.

cs.NI↗

arXiv2017

Beyond WYSIWYG: Sharing Contextual Sensing Data Through mmWave V2V Communications

In vehicular scenarios context awareness is a key enabler for road safety. However, the amount of contextual information that can be collected by a vehicle is stringently limited by the sensor technology itself (e.g., line-of-sight, coverage, weather robustness) and by the low bandwidths offered by current wireless vehicular technologies such as DSRC/802.11p. Motivated by the upsurge of research around millimeter-wave vehicle-to-anything (V2X) communications, in this work we propose a distributed vehicle-to-vehicle (V2V) association scheme that considers a quantitative measure of the potential value of the shared contextual information in the pairing process. First, we properly define the utility function of every vehicle balancing classical channel state and queuing state information (CSI/QSI) with context information i.e., sensing content resolution, timeliness and enhanced range of the sensing. Next we solve the problem via a distributed many-to-one matching game in a junction scenario with realistic vehicular mobility traces. It is shown that when receivers are able to leverage information from different sources, the average volume of collected extended sensed information under our proposed scheme is up to 71% more than that of distance and minimum delay-based matching baselines.

cs.IT↗

arXiv2016

Beamwidth Optimization in Millimeter Wave Small Cell Networks with Relay Nodes: A Swarm Intelligence Approach

Millimeter wave (mmWave) communications have been postulated as one of the most disruptive technologies for future 5G systems. Among mmWave bands the 60-GHz radio technology is specially suited for ultradense small cells and mobile data offloading scenarios. Many challenges remain to be addressed in mmWave communications but among them deafness, or misalignment between transmitter and receivers beams, and interference management lie among the most prominent ones. In the recent years, scenarios considering negligible interference on mmWave resource allocation have been rather common in literature. To this end, interestingly, many open issues still need to be addressed such as the applicability of noise-limited regime for mmWave. Furthermore, in mmWave the beam-steering mechanism imposes a forced silence period, in the course of which no data can be conveyed, that should not be neglected in throughput/delay calculations. This paper introduces mmWave enabled Small Cell Networks (SCNs) with relaying capabilities where as a result of a coordinated meta-heuristically optimized beamwidth/alignment-delay approach overall system throughput is optimized. Simulations have been conveyed for three transmitter densities under TDMA and naive 'all-on' scheduling producing average per node throughput increments of up to 248%. The paper further elaborates on the off-balancing impact of alignment delay and time-multiplexing strategies by illustrating how the foreseen transition that increasing the number of transmitters produces in the regime of a fixed-node size SCN in downlink operation fades out by a poor choice in the scheduling strategy.

cs.NI↗

arXiv2012

A Flexible Channel Coding Approach for Short-Length Codewords

This letter introduces a novel channel coding design framework for short-length codewords that permits balancing the tradeoff between the bit error rate floor and waterfall region by modifying a single real-valued parameter. The proposed approach is based on combining convolutional coding with a $q$-ary linear combination and unequal energy allocation, the latter being controlled by the aforementioned parameter. EXIT charts are used to shed light on the convergence characteristics of the associated iterative decoder, which is described in terms of factor graphs. Simulation results show that the proposed scheme is able to adjust its end-to-end error rate performance efficiently and easily, on the contrary to previous approaches that require a full code redesign when the error rate requirements of the application change. Simulations also show that, at mid-range bit-error rates, there is a small performance penalty with respect to the previous approaches. However, the EXIT chart analysis and the simulation results suggest that for very low bit-error rates the proposed system will exhibit lower error floors than previous approaches.

cs.IT↗

arXiv2012

On the Design of a Novel Joint Network-Channel Coding Scheme for the Multiple Access Relay Channel

This paper proposes a novel joint non-binary network-channel code for the Time-Division Decode-and-Forward Multiple Access Relay Channel (TD-DF-MARC), where the relay linearly combines -- over a non-binary finite field -- the coded sequences from the source nodes. A method based on an EXIT chart analysis is derived for selecting the best coefficients of the linear combination. Moreover, it is shown that for different setups of the system, different coefficients should be chosen in order to improve the performance. This conclusion contrasts with previous works where a random selection was considered. Monte Carlo simulations show that the proposed scheme outperforms, in terms of its gap to the outage probabilities, the previously published joint network-channel coding approaches. Besides, this gain is achieved by using very short-length codewords, which makes the scheme particularly attractive for low-latency applications.

cs.IT↗