Hola,
¿eres nuevo aquí?

Regístrate gratis y conecta tu empresa con financiación pública, partners y proyectos.

Tengo cuenta

Regístrate

Ver video

CNS2023-144359

Financiado

Ciencias de la computación y tecnología informática

THE MAIN GOAL OF EQBLANKETS IS TO ADVANCE AI-BASED TECHNOLOGIES TO ENABLE INTELLIGENT DISTRIBUTED COMPUTING CONTINUUM SYSTEMS (DCCS). TO THAT END, IT IS CRUCIAL TO SELECT AND ADAPT AI TECHNIQUES THAT FIT WITH THE SPECIFICITIES OF... ver más

01/01/2023

UPF

384K€

Presupuesto del proyecto: 384K€

Líder del proyecto

UNIVERSITAT POMPEU FABRA No se ha especificado una descripción o un objeto social para esta compañía.

Total investigadores 321

Financiación concedida El organismo AGENCIA ESTATAL DE INVESTIGACIÓN notifico la concesión del proyecto el día 2023-01-01 No tenemos la información de la convocatoria

0% 100%

1 Participantes

UPF

383.53K€ | Lider

Conecta tu I+D

¿Tienes un proyecto y buscas un partner? Gracias a nuestro motor inteligente podemos recomendarte los mejores socios y ponerte en contacto con ellos. Te lo explicamos en este video

Proyectos interesantes

SPINNODE SpiNNaker on the Edge 2M€ Cerrado

EIN2019-103122 PREPARACION Y OPTIMIZACION DE APLICACIONES IA A GRAN ESCALA 16K€ Cerrado

MYRTUS Multi-layer 360° dYnamic orchestrion and interopeRable desig... 6M€ Cerrado

PID2021-124463OB-I00 GESTION INTELIGENTE DEL CLOUD CONTINUUM: DESARROLLO DE LAS F... 167K€ Cerrado

RTI2018-098371-B-I00 INTELLIGENT DISTRIBUTED PROCESSING ARCHITECTURES IN FOG LEVE... 32K€ Cerrado

PID2020-116417RB-C43 TECNOLOGIAS PARA INTELIGENCIA ARTIFICIAL RECONFIGURABLE APLI... 49K€ Cerrado

Líder del proyecto

UNIVERSITAT POMPEU FABRA No se ha especificado una descripción o un objeto social para esta compañía.

Total investigadores 321

Presupuesto del proyecto 384K€

Descripción del proyecto THE MAIN GOAL OF EQBLANKETS IS TO ADVANCE AI-BASED TECHNOLOGIES TO ENABLE INTELLIGENT DISTRIBUTED COMPUTING CONTINUUM SYSTEMS (DCCS). TO THAT END, IT IS CRUCIAL TO SELECT AND ADAPT AI TECHNIQUES THAT FIT WITH THE SPECIFICITIES OF THESE TYPES OF SYSTEMS. PROVIDING FAST AND ACCURATE PREDICTIONS WHILE GIVING THE CAPACITY TO INTERPRET THE RESULTS FOR AUDIT PURPOSES AND LETTING THE SYSTEM LEARN HOW TO IMPROVE ITS EFFICIENCY AND SUSTAINABILITY. EQBLANKETS AIMS TO INTEGRATE AI PREDICTING TECHNIQUES INTO A HOLISTIC METHODOLOGY, SUCH AS DEEP NEURAL NETWORKS FOR RESOURCE UTILIZATION PREDICTION, CAUSAL INFERENCE FOR ANALYSIS AND EXPLAINABILITY PURPOSES, AND ACTIVE INFERENCE AS THE MODEL TO LEARN FROM ITS DECISIONS AND ENHANCE THE SYSTEM ACTION CAPACITY. DCCS ARE EXECUTED ON MULTIPLE TIERS COMPRISING THE INTERNET OF THINGS (IOT), EDGE, FOG, AND CLOUD INFRASTRUCTURES CONCURRENTLY. WE REFER TO SUCH SYSTEMS AS THE COMPUTING CONTINUUM (CC). EACH TIER HAS DIFFERENT BENEFITS AND LIMITATIONS; HENCE, ANY SYSTEM REQUIRES TO PROPERLY ADJUST ITS DESIGN AND MANAGEMENT STRATEGIES TO TAKE THE BEST OUT OF THE COMPUTING CONTINUUM. CONSEQUENTLY, DCCS ARE SIGNIFICANTLY DIFFERENT FROM INITIAL INTERNET SYSTEMS, WHICH WERE COMPOSED OF A SERVER AND A CLIENT AND WHERE EVERYTHING WAS PERFECTLY SPECIFIED THROUGH SOFTWARE. NOW, THEY ARE AS DEPENDENT ON THE UNDERLYING SOFTWARE INFRASTRUCTURE ON WHICH THEY ARE EXECUTED, AS ON THE LINES OF CODE THAT DEFINE THEM. IN THIS SENSE, ONE CAN IMAGINE THAT FUTURE DEVELOPMENTS WILL INCREASINGLY INTEGRATE APPLICATIONS WITH THE INFRASTRUCTURE SO THAT THE APPLICATION AND THE INFRASTRUCTURE BLEND SEAMLESSLY. REMARKABLY, THE DCCS ARE ALLOWING THE DEVELOPMENT OF KEY APPLICATIONS FOR OUR SOCIETY: SMART CITIES, SMART MANUFACTURING, AUTONOMOUS VEHICLES, HEALTHCARE, BANKING, AND MANY OTHER FIELDS ARE EAGER TO TAKE FULL ADVANTAGE OF THE PROMISES THAT THESE SYSTEMS PROVIDE: LOW LATENCY, HIGH COMPUTING CAPACITY, MOBILITY, UBIQUITY, PRIVACY, AND SECURITY.THE EQBLANKETS METHODOLOGY IS NOVEL, HOLISTIC, AND GENERIC. ON THE ONE SIDE, THE CARTESIAN REPRESENTATION ALLOWS THE INTRODUCTION OF THE SYSTEMS NEW CONTEXT AND REQUIREMENTS. THERE, THE STATE OF THE SYSTEM IS EXPRESSED IN A FRAME COMPOSED OF THE TRIPLE RESOURCES, QUALITY, AND COST (RQC). THEN, THE SYSTEM LEVERAGES SHEAF THEORY TO FORMALLY COMBINE HETEROGENEOUS SYSTEM DATA AND COMPONENTS (WP1). THIS FORMALIZATION OF THE SYSTEM ALLOWS THE CREATION OF A HIERARCHICAL REPRESENTATION OF THE SYSTEM, WHICH BY USING THE SYSTEMS DATA, IT BUILDS THE SYSTEM AS A SET OF MARKOV BLANKETS (MB). FURTHER, SLOS ARE ENVISIONED AT THE CENTER OF EACH MB TO ALLOW SYSTEM MONITORING, ANALYSIS, AND GOVERNANCE (WP2). THEN, EQBLANKETS BRINGS INTELLIGENCE TO THE SYSTEM BY USING AI-BASED METHODS, SUCH AS DEEP LEARNING OR CAUSAL INFERENCE, WHICH PRODUCES ACCURATE PREDICTIONS WHILE PROVIDING EXPLAINABILITY MEANS. FURTHER, IN TERMS OF ADAPTATION CAPABILITIES ACTIVE INFERENCE, FROM THE FEP IS AIMED TO BE USED TO ALLOW A TIMELY ADAPTATION WHILE CONTINUOUSLY IMPROVING THE AI MODELS (WP3). THE LAST STEP OF THE EQBLANKETS METHODOLOGY IS TO DEVELOP THE SYSTEMS COMPOSABILITY. THE PREVIOUSLY USED TECHNIQUES, I.E., MARKOV BLANKETS AND SHEAF THEORY, SHALL ALLOW THE SYSTEM TO BE COMPOSABLE, HENCE, EQBLANKETS WILL DEVELOP THE SPECIFIC FRAMEWORK TO ALLOW SUCH BEHAVIOR (WP4). IA\INFERENCIA ACTIVA\CAUSALIDAD\EXPLICABLE\MARKOV BLANKETS\SISTEMAS DE COMPUTACION CONTINUA DISTRIB\APRENDIZAJE AUTOMATICO

Conecta tu I+D

Entra hoy

¿Olvidó su contraseña?

Financiación

Empresas

CTIs/Universidades

Proyectos

Investigadores