Main research lines:
Data management
This line focuses on the development of data management techniques that allow addressing the challenges of the need to process and exploit large volumes of heterogeneous data in complex systems. It includes tasks such as the following: development of data management techniques for highly-dynamic mobile computing environments (vehicle networks, mobile users in a city, etc.), and with support for peer-to-peer (P2P) mobile networks; design of strategies for the analysis of large volumes of data for the development of recommendation systems to help users, in particular applied to mobile computing and smart cities (for example, tourism); and development of techniques and/or models for the exploitation of information to support decision-making in complex environments with uncertainty, where the information available may be incomplete or imprecise.
Formal Models for Complex Systems: Modeling, Analysis and Synthesis
The objective of this line is focused on the use of formalisms and the development of formal methods for the modeling, analysis (qualitative and quantitative) and synthesis in the design of complex systems of different application domains. The models built mostly belong to the family of Petri nets, although other models such as Markov chains, Transition Systems or Process Algebras are also considered.
Simulation and Execution Architectures
This line focuses on the simulation of scenarios and models to evaluate, design, make decisions or reconfigure complex systems, as well as the design of execution architectures on the cloud, which support simulations and data exploitation applications independently of the technology and the specific execution platform. It includes tasks such as the following: construction of simulators (and simulation engines) for execution in the cloud; simulation planning and analysis; construction of realistic simulation scenarios; development of a middleware that implements an architectural solution independent of commercial technologies and oriented to applications to be used in the life cycle of the complex systems considered; efficient management of computational resources (virtual machines, containers, etc.) in applications built on the cloud; and architectural proposals to combine off-line and on-line processing and guarantee service levels on cloud infrastructures.
Main active projects:
-
Title in English: Next-gEnerATion dAta Management to foster suitable Behaviors and the resilience of cItizens against modErN ChallEnges (NEAT-AMBIENCE).
In this project, we tackle data management issues to help citizens face modern challenges. It is motivated by the fact that the development of suitable data management techniques to help citizens in their daily life is more important than ever, as the world-wide context is imposing a high penalty on the quality of life of citizens. We believe that, by designing appropriate data management techniques, we can contribute to a better daily life for people and also to foster suitable behaviors in such a way that the existing modern challenges that we face as a society can be managed. We will tackle the development of solutions that adapt naturally to the distributed nature of the environment, based not only on traditional optimization metrics but also taking into account other aspects that are particularly important in our modern society, like sustainability, ecofriendliness and fairness; thus, for example, we will consider the energy consumption, the efficient exploitation of the available resources (servers, mobile devices, the edge and the cloud, etc.), and how using the proposed techniques in some scenarios could lead to a better use of the natural resources and a reduction of pollution, while maximizing the efficiency and satisfaction of users.
Project reference: PID2020-113037RB-I00.
Acknowlegment: PID2020-113037RB-I00 / AEI / 10.13039/501100011033.
Métodos para el diagnóstico y pronóstico asistidos por computador de enfermedades neurodegenerativas mediante anatomía computacional, genética en imagen y deep-learning
Title in English: Methods for the diagnosis, computer-aided prognosis of neurodegenerative diseases using computational anatomy, image genetics and deep-learning.
Neurodegenerative diseases represent a large group of neurological disorders with heterogeneous clinical and pathological expressions affecting specific subsets of neurons in specific functional anatomic systems. Research attention has been focused on Alzheimers, Parkinsons, Huntington, and amyotrophic lateral sclerosis. This is probably because these diseases are more frequent, arise for still unknown reasons, and progress in a relentless manner towards a devastating cognitive condition. Research against neurodegeneration is approached working in the development of ways to diagnose the cause of the neurodegeneration as early as possible. Regretfully, the development of effective preventive or protective therapies has been impeded by the limitations of our knowledge of the causes and the mechanisms underlying neurodegenerative diseases. We need predictive biomarkers based on imaging and genetics together with accurate predictive models of the rates of cognitive decline in those who exhibit preclinical, prodromal, or clinical disease. The purpose of this project is to develop computational tools for computer-aided diagnosis and prognosis of neurodegenerative diseases. This project focuses on the development of computational techniques useful in the quest of predictive biomarkers, the selection of the kind of neurodegeneration given the current patient condition, and the prediction of questions important in the assessment of disease evolution. These problems are approached using deep-learning with a focus on the comparison with conventional machine learning achievements and the interpretability of the models. In this project, we move towards personalized medicine arena, with the objective of finding reliable and stable biomarkers that, combined with powerful computational systems, will provide high sensitivity and specificity in single individuals towards the creation of patient-specific profiles for a precise assessment of the risk of disease onset, disease evolution, and response to treatment.
Project reference: PID2019-104358RB-I00.
Modelado y Simulación para la Ingeniería de Sistemas Discretos Complejos: Una Aproximación Basada en la Simulación Distribuida de Redes de Petri en Cloud
Title in English: Modeling and Simulation for Complex Discrete Event Engineering: An Approach Based on the Distributed Simulation of Petri Nets in Cloud
Modeling and simulation play a fundamental role in the development of system engineering processes, facilitating their design, the evaluation of architectural solutions, complementing tests, and allowing the evaluation of the system's performance. These techniques are very useful with the system already operating, to dynamically allow the redesign, configuration, monitoring and continuous maintenance of the capacity, quality and efficiency of the operational processes. In recent years, discrete event systems from fields as diverse as IoT, logistics, fleets of electric vehicles, etc., have given rise to a growing need for simulation tools for the different phases of their life cycle. All of them are of a high economic impact, of a high complexity, and highly scalable. The simulation of these systems represents, therefore, an important conceptual and technological challenge that guides to: (1) the use of formal models for their description and information extraction with the aim of generating the simulator; (2) the development of technical concepts and distributed simulation tools to make the simulation feasible, taking advantage, for example, from the concurrency inherent to the simulation of these systems through dynamic load balancing; and (3) the implementation of distributed simulations as a service on large computing infrastructures, with a preference for cloud-like platforms. These are the objectives of this project. The validation of the proposals will be carried out on realistic systems extracted from case studies in fields such as logistics, fleets of electric vehicles for recharging, or cloud systems themselves.
Project reference: PGC2018-099815-B-I00.
AI4Europe
Title in English: An AI On-Demand Platform to Support Research Excellence in Europe.
AI scientists and researchers are required to invest a lot of effort to identify trustworthy, high-quality datasets, physical resources, algorithms, and find efficient mechanisms to communicate, cooperate, and engage in an open and transparent manner. For the European AI strategy to succeed, methods to unite the paradigms of AI research, application, and data are needed. It is also essential that these solutions follow the European seal of quality ensuring trustworthy and explainable methods.
To foster an Ecosystem of Excellence and to accelerate the adoption of solutions based on AI, the work of the AI4Europe project will overcome these challenges extending, and improving the existing AI ecosystem to incorporate and improve the use from the research, academic and innovation community, while continuing to support SMEs and connected projects.
This evolution requires structures to support easily reproducible and open access results and outputs. Research on AI — and AI in research and science — must be transferable to foster innovation, reducing the gap between industrial research and academic research, and with special attention on social impact, and geographical and individual human diversity. To succeed, the project must facilitate openness and transparent access to AI across a spectrum of stakeholders from students of AI to innovators, provide them with tools that ease their needs and empower their understanding and research on AI, and to incorporate their needs into the roadmap of the platform as they are the main supporters of its future.
The roles of the University of Zaragoza within the project are, first, to design and lead the AI4Europe Reproducibility of Experiments Strategy and to coordinate its implementation. Second, to provide support in the design and implementation aspects of distributed computing needs for AI4Europe.
Project reference: Program Horizon, topic identification HORIZON-CL4-2021-HUMAN-01-02, project reference 101070000.
Acknowledgment: Horizon AI4Europe - 101070000.
NEAT-AMBIENCESome past projects:
Datos 4.0: Retos y Soluciones
Title in English: Data 4.0: Challenges and Solutions.
This project focuses on challenges and solutions for what we call Data 4.0: the fourth revolution in data management, which in addition to being "big" and/or "smart" requires new processing solutions and exploitation in demanding scenarios of a whole new range of applications (unthinkable a few years ago) in scenarios of different nature. The global project is a coordinated project between 4 universities (University of A Coruña, Polytechnic University of Madrid, University of the Basque Country and University of Zaragoza). In particular, at the node of the University of Zaragoza we focus on the area of mobile computing, fundamentally addressing the following 2 challenges: 1) data exploitation in mobile environments, which involves obtaining and exploiting useful information in wireless computing contexts by evaluating the relevance of the data produced and filtering them based on the user's context (location, activity being carried out, etc.); and 2) management of the semantic heterogeneity of the data, which implies creating mechanisms that discover and make the meaning of the data 4.0 explicit, that is, that represent the real meaning of the accessible data, help users to express the type of information they seek, and consider those meanings to answer queries. The ultimate goal is to provide users with exactly the relevant information they need.
Project reference: TIN2016-78011-C4-3-R.
TRAFAIR - Understanding Traffic Flows to Improve Air quality
TRAFAIR raises awareness among citizens and public administrations about the air quality within an urban environment and the pollution caused by traffic. The project aims at monitoring air quality by using sensors in 6 cities and making air quality predictions thanks to simulation models. The two main goals of the project are: 1) monitoring urban air quality by using sensors in 6 European cities: Zaragoza (600,000 inhabitants), Florence (382,000), Modena (185,000), Livorno (160,000), Santiago de Compostela (95,000) and Pisa (90,000); and 2) making urban air quality predictions thanks to simulation models based on weather forecast and traffic flows.
Monitoring air quality means to set up a network of low-cost sensors spread within the city to monitor levels of pollutions in areas that are not covered by the legal air quality stations.
Predicting urban air quality is possible thanks to a chain of simulation models. The traffic flows are simulated from the real measurements supplied by traffic sensors, then the emissions are calculated by taking into account the vehicle fleet in the city. Finally, the air quality predictions are calculated using an air pollution dispersion model taking into account the emissions, building shapes, and weather forecast.
Project in cooperation among several research groups (led by Raquel Trillo at the University of Zaragoza).
Project reference: 2017-EU-IA-0167.
