Edge artificial intelligence (AI) combines AI, the Internet of Things (IoT) and edge computing technologies to provide real-time data collection, processing, analytics, and decision-making. 

Cutting-edge AI, generative AI, agentic AI technologies and applications are constantly evolving and driving intelligent control and decisions of the future by moving AI capabilities closer to the data source from the physical world to enable the implementation of autonomous systems.

Advancements in smart sensors, edge processing devices with novel processing architectures, artificial and spiking neural networks implemented through CPUs, GPUs, NPUs, TPUs, ASICs, FPGAs, system-on-chip (SoC), system-on-module (SoM), and neuromorphic accelerators support the development of edge AI.

The expansion of the micro-, deep-, and meta-edge continuum as part of the novel distributed AI-based computing architectures supports the emerging intelligent autonomous systems and applications across various industrial sectors.

The European Conference on Edge AI Technologies and Applications – EEAI brings together researchers, practitioners, and edge intelligence champions to convey and exemplify their perspectives on energy-efficient edge AI architectures, frameworks, platforms and applications.

The conference addresses the edge AI foundations, applications, algorithms, hardware, software, communication networks, and systems. It showcases the latest advancements in these areas and provides an arena for debating and identifying future directions and challenges.

The European Conference on Edge AI Technologies and Applications aims to serve as an innovative edge AI platform for exchanging ideas and concepts. It will feature several leading representatives in research and industry who support the integration of edge AI technologies to advance future industrial applications.

The synergy and convergence of technologies enables us to envision systems that not only process information but also understand context, generate innovative solutions, and execute complex tasks directly at the source. This is the next step of digital transformation, the moment we transition from creating tools that analyse the world to building intelligent agents that can act within it. It’s a future where intelligent edge devices respond to commands and anticipate needs, where technology doesn’t just present data but creates art, and where autonomous systems work alongside us to solve humanity’s most significant challenges.

In this context, the topics of interest include, but are not limited to, the followings:

Edge AI Technologies

• Edge AI tools and methods
• AI hardware-software co-design methods and tools
• Edge AI workflows and systems design
• Edge AI auto ML
• Architectures, frameworks, and protocols for intelligent edge processing
• Machine vision: image classification, object detection, semantic segmentation
• Edge AI natural language processing, speech recognition and synthesis
• Energy optimisation for edge devices
• Reconfigurable edge AI
• Optimisation methods and techniques for neural networks
• Ultra-low-bit quantization and binarization
• Hardware-aware neural architecture search (NAS) and pruning
• Energy-aware edge AI and power-adaptive inference
• On-device continual learning
• On-device modelling and predictive control
• Multimodal learning implications for edge AI training and inference
• Federated learning
• Advanced federated learning and swarm intelligence
• Edge-native reinforcement learning
• Multi-agent collaboration at the edge
• Edge AI accelerators – GPUs, NPUs, TPUs, ASICs, FPGAs, etc.
• Neuromorphic computing for ultra-low-power edge AI
• Neuromorphic HW Architectures (digital, mixed signal, analog, spiking)
• RISC-V-based edge AI hardware acceleration
• Edge AI heterogeneous systems integration
• Heterogeneous edge AI: integrating CPUs, GPUs, NPUs, and TPUs in edge devices to optimize AI workloads
• Compiler and firmware co-design for edge AI workloads
• Smart connectivity at the edge
• Edge AI optimisation for 5G/6G
• Edge AI for immersive technologies (XR/VR/AR/MR, metaverse/omniverse/multiverse)
• Enhancing immersive experiences with real-time AI processing at the edge
• Simulation and analysis techniques for edge intelligence
• Human to edge AI interaction and its implications for privacy
• Integrated hardware and software edge platforms
• Agentic and embodied AI at the edge
• Generative edge AI
• Compact and efficient generative architectures
• Frameworks for generative and agentic edge AI
• Generative data augmentation for edge model training
• Transformers at the edge: data, hardware, and software engineering challenges
• Sustainable edge AI technologies
• Trustworthy edge AI (dependable edge AI systems – security, privacy, safety reliability, availability, maintainability, connectability, fairness, accountable, robustness, resilience)
• Adversarial robustness and security foe edge AI
• Edge AI verification, validation and testing
• Safe, verifiable, explainable, interpretable AI for autonomous edge agents
• Benchmarking edge AI models
• Edge AI explainability and interpretability
• Ethical implications of edge AI

Edge AI system holistic approach: converging hardware, software, AI technology stack and data

Ovidiu Vermesan, SINTEF, Norway

Marcello Coppola, STMicroelectronics, France      

Carmelo Pino, STMicroelectronics, Italy

Patrick Pype, NXP Semiconductors, Belgium

Alain Pagani, German Research Center for Artificial Intelligence – DFKI, Germany

Luca Valcarenghi, Sant’Anna School of Advanced Studies, Italy

Konstantinos Georgopoulos, Technical University of Crete, Greece

Iakovos Mavroidis, Technical University of Crete, Greece

Huascar Espinoza, Key Digital Technologies Joint Undertaking, Belgium

Paolo Azzoni, Inside Industry Association, Netherlands

Inessa Seifert, EPoSS – European Platform of System Integrators, Germany

Björn Debaillie, imec, Belgium

Fabian Chersi, CEA-LIST, France

Hana Krichene, CEA-LIST, France

Pavel Smrz, COGNITECHNA s.r.o., Czech Republic

Andrea Dunbar, CSEM, Switzerland

Jochen Koszescha, Infineon, Germany

George Dimitrakopoulos, Infineon, Germany

Holger Schmidt, Infineon, Germany

Roy Bahr, SINTEF, Norway

Ihsen Alouani, Université Polytechnique Hauts-De-France, France

Modris Greitans, Institute of Electronics and Computer Science, Latvia

Janis Arents, Institute of Electronics and Computer Science, Latvia

Andrea Acquaviva, University of Bologna, Italy

Francesco Barchi, University of Bologna, Italy

Paolo Meloni, University of Cagliari, Italy

Fetze Pijlman, Signify, Netherlands

Vincenzo Piuri, University of Milan, Italy

Fabio Scotti, University of Milan, Italy

Angelo Genovese, University of Milan, Italy

Marco Ottella, Xtremion Engineering Srls, Italy

Gleb Radchenko, Silicon Austria Labs, Austria

Alberto Faro, DeepSensing, Italy