DAIS - Distributed Artificial Intelligent Systems
Divisions:
Status:
Start date:
The use of artificial intelligence (AI) in Edge computing is entering a new era based on the use of ubiquitous small and connected devices. Until now, Europe has not been doing well, as America sets the standards and most components are produced in Asia or America. This project believes doing better is realized by (1) Putting European values of selforganization, privacy by design and low use of energy in the core of the Edge Computing components that shape this new era, and delivering the technology needed to promote these values; (2) Focusing on pan European cooperation to ramp up the capabilities needed to deliver these new components at a scale that can make a real impact. Europe does not have huge IT leaders so cooperation from a very early phase is key. All partners in the project participate in delivering key parts of these new Edge Computing components; and (3) Demonstrating the use of these components in key European industrial areas. Clear and early examples are needed to un-lock corporate and external funding to deliver on the promise of this very exciting project. The DAIS project will research and deliver distributed artificial intelligent systems. It will not research new algorithms, as such, but solves the problems of running existing algorithms on these vastly distributed edge devices that are designed based on the above three European core values. The research and innovation activities are organized around eight complementary and mutually supportive supply chains. Five of these focus on delivering the hardware and software that is needed to run industrial-grade AI on different types of networking topologies. Three of the supply chains demonstrate how known AI challenges, from different functional areas, are met by this pan European effort. The DAIS project consists of 48 parties from 11 different countries. The DAIS project fosters cooperation between large and leading industrial players from different domains.
Network intrusion detection using machine learning on resource-constrained edge devices (Jul 2024) Pontus Lidholm , Tijana Markovic, Miguel Leon Ortiz, Per Erik Strandberg International Conference on Neural Networks (IJCNN 24)
Random forest with differential privacy in federated learning framework for network attack detection and classification (Jun 2024) Tijana Markovic, Miguel Leon Ortiz, David Buffoni , Sasikumar Punnekkat Applied Intelligence (APIN)
Balancing Privacy and Performance in Federated Learning: a Systematic Literature Review on Methods and Metrics (Mar 2024) Samaneh Mohammadi, Ali Balador, Sima Sinaei, Francesco Flammini Journal of Parallel and Distributed Computing (JPDC)
Hyperparameters Optimization for Federated Learning System: Speech Emotion Recognition Case Study (Oct 2023) Kateryna Mishchenko , Samaneh Mohammadi, Mohammadreza Mohammadi , Sima Sinaei The Eighth IEEE International Conference on Fog and Mobile Edge Computing (FMEC 2023)
Balancing Privacy and Accuracy in Federated Learning for Speech Emotion Recognition (Sep 2023) Samaneh Mohammadi, Mohammadreza Mohammadi , Sima Sinaei, Ali Balador, Ehsan Nowroozi , Francesco Flammini, Mauro Conti 18th Conference on Computer Science and Intelligence Systems (FedCSIS 2023)
Optimized Paillier Homomorphic Encryption in Federated Learning for Speech Emotion Recognition (Aug 2023) Samaneh Mohammadi, Sima Sinaei, Ali Balador, Francesco Flammini IEEE 47th Annual Computers, Software, and Applications Conference (COMPSAC 2023)
