Distributed and Ubiquitous Intelligent Systems: HPC, Edge Computing and Big Data
This line of research uses specialised hardware for the construction of intelligent systems that can later be used in real world applications that work in changing environments and with uncertainty, that use large amounts of data and have major temporary restrictions. To resolve these complex optimisation and learning tasks we use different platforms: multiprocessors, equipment clusters, GPUs and even smartphones and raspberry-pi. We work both on centralized processed where data flow towards the computational resources and on Edge/Fog computing, which is where the information I obtained or where the end user waits for the results. We provide tools for crowd-computing on web browsers, voluntarily resolving social impact problems on end user devices. We also focus on new techniques for highly efficient learning in a computational laboratory. Along this line we also address aspects related with intelligent vehicles, communications between vehicles (V2V) and the use of micro-simulations for computing and communication systems.