arXiv subjects

Ana Fernández-Vilas

Publications and source records attributed to Ana Fernández-Vilas.

29 records · Page 2Linked to original sources

arXiv2023

Decentralised and collaborative machine learning framework for IoT

Decentralised machine learning has recently been proposed as a potential solution to the security issues of the canonical federated learning approach. In this paper, we propose a decentralised and collaborative machine learning framework specially oriented to resource-constrained devices, usual in IoT deployments. With this aim we propose the following construction blocks. First, an incremental learning algorithm based on prototypes that was specifically implemented to work in low-performance computing elements. Second, two random-based protocols to exchange the local models among the computing elements in the network. Finally, two algorithmics approaches for prediction and prototype creation. This proposal was compared to a typical centralized incremental learning approach in terms of accuracy, training time and robustness with very promising results.

cs.LG↗

arXiv2023

Collaboratively assessing urban alerts in ad hoc participatory sensing

Ad hoc architectures have emerged as a valuable alternative to centralized participatory sensing systems due to their infrastructureless nature, which ensures good availability, easy maintenance and direct user communication. As a result, they need to incorporate content-aware assessment mechanisms to deal with a common problem in participatory sensing: information assessment. Easy contribution encourages users participation and improves the sensing task but may result in large amounts of data, which may not be valid or relevant. Currently, prioritization is the only totally ad hoc scheme to assess user-generated alerts. This strategy prevents duplicates from congesting the network. However, it does not include the assessment of every generated alert and does not deal with low-quality or irrelevant alerts. In order to ensure users receive only interesting alerts and the network is not compromised, we propose two collaborative alert assessment mechanisms that, while keeping the network flat, provide an effective message filter. Both of them rely on opportunistic collaboration with nearby peers. By simulating their behavior in a real urban area, we have proved them able to decrease network load while maintaining alert delivery ratio.

cs.NI↗

arXiv2023

Fog Architectures and Sensor Location Certification in Distributed Event-Based Systems

Since smart cities aim at becoming self-monitoring and self-response systems, their deployment relies on close resource monitoring through large-scale urban sensing. The subsequent gathering of massive amounts of data makes essential the development of event-filtering mechanisms that enable the selection of what is relevant and trustworthy. Due to the rise of mobile event producers, location information has become a valuable filtering criterion, as it not only offers extra information on the described event, but also enhances trust in the producer. Implementing mechanisms that validate the quality of location information becomes then imperative. The lack of such strategies in cloud architectures compels the adoption of new communication schemes for Internet of Things (IoT)-based urban services. To serve the demand for location verification in urban event-based systems (DEBS), we have designed three different fog architectures that combine proximity and cloud communication. We have used network simulations with realistic urban traces to prove that the three of them can correctly identify between 73% and 100% of false location claims.

cs.NI↗

arXiv2023

Classification of retail products: From probabilistic ranking to neural networks

Food retailing is now on an accelerated path to a success penetration into the digital market by new ways of value creation at all stages of the consumer decision process. One of the most important imperatives in this path is the availability of quality data to feed all the process in digital transformation. But the quality of data is not so obvious if we consider the variety of products and suppliers in the grocery market. Within this context of digital transformation of grocery industry, \textit{Midiadia} is Spanish data provider company that works on converting data from the retailers' products into knowledge with attributes and insights from the product labels, that is, maintaining quality data in a dynamic market with a high dispersion of products. Currently, they manually categorize products (groceries) according to the information extracted directly (text processing) from the product labelling and packaging. This paper introduces a solution to automatically categorize the constantly changing product catalogue into a 3-level food taxonomy. Our proposal studies three different approaches: a score-based ranking method, traditional machine learning algorithms, and deep neural networks. Thus, we provide four different classifiers that support a more efficient and less error-prone maintenance of groceries catalogues, the main asset of the company. Finally, we have compared the performance of these three alternatives, concluding that traditional machine learning algorithms perform better, but closely followed by the score-based approach.

cs.AI↗

arXiv2023

Multi-criteria recommendation systems to foster online grocery

With the exponential increase in information, it has become imperative to design mechanisms that allow users to access what matters to them as quickly as possible. The recommendation system ($RS$) with information technology development is the solution, it is an intelligent system. Various types of data can be collected on items of interest to users and presented as recommendations. $RS$ also play a very important role in e-commerce. The purpose of recommending a product is to designate the most appropriate designation for a specific product. The major challenges when recommending products are insufficient information about the products and the categories to which they belong. In this paper, we transform the product data using two methods of document representation: bag-of-words (BOW) and the neural network-based document combination known as vector-based (Doc2Vec). We propose three-criteria recommendation systems (product, package, and health) for each document representation method to foster online grocery, which depends on product characteristics such as (composition, packaging, nutrition table, allergen, etc.). For our evaluation, we conducted a user and expert survey. Finally, we have compared the performance of these three criteria for each document representation method, discovering that the neural network-based (Doc2Vec) performs better and completely alters the results.

cs.IR↗

arXiv2023

Integrating micro-learning content in traditional e-learning platforms

Lifelong learning requires appropriate solutions, especially for corporate training. Workers usually have difficulty combining training and their normal work. In this context, micro-learning emerges as a suitable solution, since it is based on breaking down new concepts into small fragments or pills of content, which can be consumed in short periods of time. The purpose of this paper is twofold. First, we offer an updated overview of the research on this training paradigm, as well as the different technologies leading to potential commercial solutions. Second, we introduce a proposal to add micro-learning content to more formal distance learning environments (traditional Learning Management Systems or LMS), with the aim of taking advantage of both learning philosophies. Our approach is based on a Service-Oriented Architecture (SOA) that is deployed in the cloud. In order to ensure the full integration of the micro-learning approach in traditional LMSs, we have used two well-known standards in the distance learning field: LTI (Learning Tools Interoperability) and LIS (Learning Information Service). The combination of these two technologies allows the exchange of data with the LMS to monitor the student's activity and results. Finally, we have collected the opinion of lectures from different countries in order to know their thoughts about the potential of this new approach in higher education, obtaining positive feedback.

cs.NI↗

arXiv2023

Irregular Repetition Slotted Aloha with Multipacket Detection: A Density Evolution Analysis

Irregular repetition slotted Aloha (IRSA) has shown significant advantages as a modern technique for uncoordinated random access with massive number of users due to its capability of achieving theoretically a throughput of $1$ packet per slot. When the receiver has also the multi-packet reception of multi-user (MUD) detection property, by applying successive interference cancellation, IRSA also obtains very low packet loss probabilities at low traffic loads, but is unable in general to achieve a normalized throughput close to the $1$. In this paper, we reconsider the case of IRSA with $k$-MUD receivers and derive the general density evolution equations for the non-asymptotic analysis of the packet loss rate, for arbitrary frame lengths and two variants of the first slot used for transmission. Next, using the potential function, we give new capacity bounds on the capacity of the system, showing the threshold arrival rate for zero decoding error probability. Our numerical results illustrate performance in terms of throughput and average delay for $k$-MUD IRSA with finite memory at the receiver, and also with bounded maximum delay.

cs.IT↗

arXiv2023

KPIs-Based Clustering and Visualization of HPC jobs: a Feature Reduction Approach

High-Performance Computing (HPC) systems need to be constantly monitored to ensure their stability. The monitoring systems collect a tremendous amount of data about different parameters or Key Performance Indicators (KPIs), such as resource usage, IO waiting time, etc. A proper analysis of this data, usually stored as time series, can provide insight in choosing the right management strategies as well as the early detection of issues. In this paper, we introduce a methodology to cluster HPC jobs according to their KPI indicators. Our approach reduces the inherent high dimensionality of the collected data by applying two techniques to the time series: literature-based and variance-based feature extraction. We also define a procedure to visualize the obtained clusters by combining the two previous approaches and the Principal Component Analysis (PCA). Finally, we have validated our contributions on a real data set to conclude that those KPIs related to CPU usage provide the best cohesion and separation for clustering analysis and the good results of our visualization methodology.

cs.AI↗

arXiv2023

Unsupervised KPIs-Based Clustering of Jobs in HPC Data Centers

Performance analysis is an essential task in High-Performance Computing (HPC) systems and it is applied for different purposes such as anomaly detection, optimal resource allocation, and budget planning. HPC monitoring tasks generate a huge number of Key Performance Indicators (KPIs) to supervise the status of the jobs running in these systems. KPIs give data about CPU usage, memory usage, network (interface) traffic, or other sensors that monitor the hardware. Analyzing this data, it is possible to obtain insightful information about running jobs, such as their characteristics, performance, and failures. The main contribution in this paper is to identify which metric/s (KPIs) is/are the most appropriate to identify/classify different types of jobs according to their behavior in the HPC system. With this aim, we have applied different clustering techniques (partition and hierarchical clustering algorithms) using a real dataset from the Galician Computation Center (CESGA). We have concluded that (i) those metrics (KPIs) related to the Network (interface) traffic monitoring provide the best cohesion and separation to cluster HPC jobs, and (ii) hierarchical clustering algorithms are the most suitable for this task. Our approach was validated using a different real dataset from the same HPC center.

cs.DC↗

arXiv2023

Scheduling and Communication Schemes for Decentralized Federated Learning

Federated learning (FL) is a distributed machine learning paradigm in which a large number of clients coordinate with a central server to learn a model without sharing their own training data. One central server is not enough, due to problems of connectivity with clients. In this paper, a decentralized federated learning (DFL) model with the stochastic gradient descent (SGD) algorithm has been introduced, as a more scalable approach to improve the learning performance in a network of agents with arbitrary topology. Three scheduling policies for DFL have been proposed for communications between the clients and the parallel servers, and the convergence, accuracy, and loss have been tested in a totally decentralized mplementation of SGD. The experimental results show that the proposed scheduling polices have an impact both on the speed of convergence and in the final global model.

cs.LG↗

arXiv2023

Using Decentralized Aggregation for Federated Learning with Differential Privacy

Nowadays, the ubiquitous usage of mobile devices and networks have raised concerns about the loss of control over personal data and research advance towards the trade-off between privacy and utility in scenarios that combine exchange communications, big databases and distributed and collaborative (P2P) Machine Learning techniques. On the other hand, although Federated Learning (FL) provides some level of privacy by retaining the data at the local node, which executes a local training to enrich a global model, this scenario is still susceptible to privacy breaches as membership inference attacks. To provide a stronger level of privacy, this research deploys an experimental environment for FL with Differential Privacy (DP) using benchmark datasets. The obtained results show that the election of parameters and techniques of DP is central in the aforementioned trade-off between privacy and utility by means of a classification example.

cs.LG↗