Artificial Inteligence

Bringing new and faster AI methods to
unconventional emerging applications

The AI research group is devoted not only to the application of advanced machine learning techniques in new relevant applications related to nanonetworking, but also to the development of novel acceleration techniques for emerging AI algorithms. As such, the group works in close collaboration to the rest of the groups of the center; with CHIPS to develop better hardware accelerators, with HUMAN to develop localization techniques for intra-body networks or to improve several medical applications related with Electroencephalogram (EEG) analysis, and with QUANTUM applying AI to develop better compilation techniques for quantum computers. 

Albert Cabellos

Group Leader

Prof. Albert Cabellos, long-time scientific director of N3Cat and also founder of the Barcelona Neural Networking (BNN) Lab, brings his unique expertise to lead the AI group in N3Cat. Pioneer of the application of novel AI techniques such as Graph Neural Networks (GNNs) or topological learning in the domain of computer networks, he shares the ambition of the group, which is to apply these novel techniques in new and exciting application areas and to develop new hardware and software acceleration techniques for these novel AI workloads.

PROJECTS

winc
WINC: Wireless Networks within Next-Generation Computing Systems
ERC Starting Grant
2022-2027
overview2
IGNNSPECTOR: Graph-Driven Acceleration of Graph Neural Networks
NEC Labs Fellowship
2021-2022
EmbeddedImage
ScaleITN: Scalable Localization-enabled In-body Terahertz Nanonetwork
MSCA Postdoctoral Fellowship
2020-2022

SELECTED PUBLICATIONS

Computing Graph Neural Networks: A Survey from Algorithms to Accelerators

S. Abadal, A. Jain, R. Guirado, J. López-Alonso, E. Alarcón

ACM Computing Surveys, 2021

Tailoring Graph Neural Network-based Flow-guided Localization to Individual Bloodstreams and Activities

P. Galván, F. Lemic, G. Calvo, S. Abadal, X. Costa-Perez

Proceedings of the ACM NANOCOM ’24

ProGNNosis: A Data-driven Model to Predict GNN Computation Time Using Graph Metrics

A. Wassington and S. Abadal

Proceedings of the HiPEAC'22 Workshops

Circuit Partitioning for Multi-Core Quantum Architectures with Deep Reinforcement Learning

A. Pastor, P. Escofet, S. Ben Rached, E. Alarcón, P. Barlet-Ros and S. Abadal

Proceedings of the IEEE ISCAS ‘24

WHYPE: A Scale-Out Architecture with Wireless Over-the-Air Majority for Scalable In-memory Hyperdimensional Computing

R. Guirado, A. Rahimi, G. Karunaratne, E. Alarcón, A. Sebastian and S. Abadal

IEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2023