Home   >   Competence centers   >   VRAIN

VRAIN

Valencian Research Institute for Artificial Intelligence

contact
Aída Gil / Ana Cidad
Address
Camino de Vera S/N
Province
Valencia

WHO WE ARE

VRAIN is a reference University Institute of Artificial Intelligence in Spain. Formed by 112 researchers from 7 research groups, members of the DSIC and the PROS Centre of the Universitat Politècnica de València, with more than 30 years of experience in different areas of AI research and referents for their scientific production at international level.

Technological Enablers Capabilities

Advanced industrial technologies

Smart manufacturing and processes
- Autonomous Computing, Self-adaptation and "Human in the loop" in industrial environments (Design and testing of Internet of Things (IoT) in enterprise environments: Integration of real world objects with information systems, Design and development of ubiquitous and ambient intelligence systems applied to Smart Homes, Buildings and Cities). Design of autonomous and self-adaptive systems) - Organizational Modeling and Requirements Engineering (From Requirements to Code, Requirements Management in Organizational Modeling, Support for Information Systems Evolution, Model and Ontology Driven Interoperability of Information Systems, Ubiquitous Business Processes and Business Process Variability Management) - AI and industry: sustainability and smart manufacturing Sustainable manufacturing is defined as "the creation of manufactured products that use processes that minimize negative environmental impacts, conserve energy and natural resources, are safe for employees, communities and consumers, and are economically affordable". The notions of clean and green manufacturing (to name a few) are often used to describe strategies or philosophies that are more or less similar in the context of industrial sustainability. In this line of research we focus on the application of artificial intelligence techniques to achieve optimized solutions for manufacturing systems in which sustainable objectives are combined with economic, temporal and other objectives. Intelligent manufacturing systems are one of the research lines of Factories of the Future and Industry 4.0, it is conceived as a distributed intelligent system in which each component, element and/or manufacturing resource is modelled and controlled by software "agents" that can cooperate to solve complex problems. Intelligent manufacturing control systems are known to meet the need for reactivity and adaptability of future sustainable manufacturing systems. Cooperation between agents allows for an equitable allocation of resources (such as energy allocation) that can lead to "social welfare". In addition, Cyber Pyshical Systems, Systems of Systems, Industrial IoT and Digital Twins are related sublines that the group is working on.

Sectores

  • Manufactura
  • Transporte, almacenamiento y logística

Artificial Intelligence and Computing

Machine learning technologies
- Automated software testing (Test automation consulting, Automated testing at user interface level www.TESTAR.org, Security testing, Evaluation and improvement of software development and testing processes following TMAP, CMMI, IEC, etc. standards). - Metamodels for Adversarial Machine Learning. - Inductive programming. - Model adaptation. - Hierarchical and Multiclass classification. - Machine Intelligence Evaluation

Sectores

  • Otros
  • Salud
Natural language processing and text mining
Automatic Speech Recognition (ASR) Machine Translation (MT) Speech synthesis. Design of acoustic and language models, entity recognition, morphosyntactic tagging for information retrieval, voice search and answer search, automatic summarization and content analysis in social networks (emotion analysis, polarity, reputation analysis). Text mining: Content analysis in social networks (analysis of feelings, emotions, reputation, trends and positioning among others), in the automatic summary of documents (text and audio) and in problems that require a semantic representation such as those related to the search for answers, double meaning detection, topic classification, etc.

Sectores

  • Otros
  • Turismo
Responsible AI
Responsible Artificial Intelligence (Responsible AI -RAI-; AI Safety/Security and Explainable AI -XAI-) Starting from the premise that artificial intelligence must be responsible, reliable, based on ethical and social values, human-centered, safe and secure and interpretable/explainable, we work in several directions. On the one hand, in relation to learning we work on techniques for explaining black box models (mimetic models), on the analysis of adversarial attacks according to the learning family used and, more recently, on how different AI paradigms are related and their effect on their security and robustness. On the other hand, in relation to symbolic methods, we work on provable security, cryptographic protocol analysis, reversible computation and formal methods for the analysis, modeling, verification, synthesis, debugging, certification, transformation, and optimization of critical and intelligent systems. We apply normative systems theory to endow AI systems with ethical values.

Sectores

  • Otros
  • Salud
Effective data processing infrastructures and platforms
AI Readiness and Automation (AutoDS, AutoAI and Technological Readiness).

Sectores

  • Hábitat
  • Otros
Affective computing
Affective Intelligent Agents. Being able to reproduce human social behaviour in complex domains (e.g. modelling fear of loss and sellers' behaviour in housing markets, predicting risk-taking in auction experiments, crisis management simulation frameworks), is of great interest in many research areas, such as emergency management, migration flows, personalised education, healthcare, video games or marketing.

Sectores

  • Hábitat
  • Salud
Multiagent Systems and Agreement technologies
Complex networks and intelligent systems. In the context of MAS and complex networks: Analysis of the influence of different types of network structures on the management of distributed resources in agent societies; Self-adaptation mechanisms in networks to improve network navigation; Influence of network structure on the emergence of cooperation in agent societies; Development of unsupervised agreements between intelligent autonomous entities using consensus processes in networks; Analysis of the evolution of structural properties in social networks. Social Simulation. Agreement Technologies (AT) deal with large-scale open distributed computing systems in which autonomous software agents negotiate with each other, usually on behalf of humans, in order to reach mutually acceptable agreements that must be consistent with the normative context in which they are set and that will allow, once accepted, agents to request and respect each other's services. These technologies include virtual organizations, automatic negotiation, normative systems, trust and reputation models, automatic argumentation, etc.

Sectores

  • Salud
  • Turismo
Planning, scheduling and recommander systems
Planning and consulting systems. Planning problems are related to problems of decision making and actions that should take place in particular environments to achieve specific objectives. Decisions will be determined by the context of the problem, the pattern of action, the timing of actions and the resources required. The multi-agent view requires cooperative planning to combine and complement the actions that different actors should take. Demographic, filtering or social context and behavioural analysis techniques are applied. Recommender systems A recommender system is a personalization tool that attempts to provide people with a list of information items that best suits their individual tastes. It influences user preferences by analyzing available user data, information about other users, and information about the environment, and selects the elements that best match these preferences. In short, a recommender system offers the possibility to customize the filtering of information so that only information adapted to the user's needs and preferences is displayed. We have a long experience in the development of tourism recommender systems, both for individuals and groups, based on different types of information, such as explicit/implicit preferences, user psychology, etc. - Autonomous vehicles. Autonomous systems in robotics are designed to perform tasks without the supervision of human controllers. There are several advantages: minimizing controller fatigue, minimizing risk, minimizing operating costs, and improving product and quality control operations.

Sectores

  • Otros
  • Transporte, almacenamiento y logística
Multiagent Systems and Agreement technologies
- Persuasion Technologies. Persuasive technologies is an interdisciplinary field of research that focuses on the design and development of interactive technologies that can create, maintain, or change human thought and behavior using persuasive techniques. These technologies are still an emerging scientific field that faces multiple challenges, such as: user acceptance and actual use of the technology; adaptation and customization of the system for the user; perceived usefulness and user satisfaction; detection of events at the right time; continuity of the new behavior over time; and ethical issues.

Sectores

  • Manufactura
  • Transporte, almacenamiento y logística
Multiagent Systems and Agreement technologies
'- Multi-agent systems. The multi-agent systems paradigm is a growing area of interest in AI. This is due to its application to the resolution of complex problems, where classical techniques fail to obtain a satisfactory solution. Multi-agent systems (MAS) address the need for communication and collaboration between autonomous agents (entities whose behaviour is autonomous) - Virtual markets.

Sectores

  • Otros
  • Suministro de energía eléctrica, gas, agua
Responsible AI
Explainable AI in the Life Sciences Understanding and manipulating the genome implies understanding and manipulating life as we perceive it on our planet. This major challenge has implications for everything related to health. This line of work aims to develop platforms that are able to search, identify, retrieve and interpret relevant genomic information. This information must be obtained from constantly growing repositories, with immense volumes of data in which machine learning techniques are indispensable to select the information of value in a context of clinical diagnosis and preventive treatment that integrates genomic knowledge into medical practice as we understand it. Integrating conventional clinical information with genomic information in an AI setting is essential to develop viable and reliable precision medicine. The application of AI techniques is essential to interpret the chaos of genomic data being generated in order to determine which data are clinically relevant.

Sectores

  • Agri-food
  • Salud
Big data analytics technologies
Educational technologies and Big Data By big data we mean a collection of massive and complex data sets that are difficult to process using conventional data management tools and applications. Educational technologies are one area where big data is gaining momentum at a rapid pace. The research is dedicated to the study and implementation of innovative solutions for the processing of massive data in the area of educational technologies. In particular, emphasis is placed on the development and integration of innovative digital tools, solutions and services for learning and teaching. Software production methods for the development of Big Data applications. Genomic Data Management

Sectores

  • Otros
  • Salud

Ciber Connectivity

Cyber-Physical Systems
Autonomic Computing, Self-adaptation and "Human in the loop" (Design and testing of Internet of Things (IoT) in enterprise environments: Integration of real-world objects with information systems, Design and development of ubiquitous and ambient intelligence systems applied to Smart Homes, Buildings and Cities). Design of autonomous and self-adaptive systems)

Sectores

  • Hábitat
  • Turismo
Advanced Human-Machine Interfaces
Human-Computer Interaction, non-intrusive interaction and usability (Usability analysis of information systems in DSDM environments, Development of multi-platform/device user interfaces) Context-sensitive/adaptive user interfaces. Ubiquitous interactions in the IoT.

Sectores

  • Hábitat
  • Turismo

Ciber Security

Data security and privacy technologies
Software specification and verification. Demonstrable safety. Automated analysis, modeling, verification, synthesis, learning, debugging and optimization of complex systems, models and data. Analysis of chrotographic protocols.

Sectores

  • Otros

R+d+i Projects

Patents

Staff

Training

Demonstrators

The Demonstrators are spaces enabled by each center to show application examples and prototypes of the Enabling Technologies

Infraestructures

Access the different infrastructures and equipment that each member makes available to companies and other centers related to each technology.