Edge AI Academy Summer School 2025 Overview
The rapid evolution of Edge Artificial Intelligence (AI) and RISC-V technologies is advancing novel computing architectures, bringing powerful AI capabilities closer to the physical world. Edge AI provides intelligent control and decision-making capabilities directly where data is generated, enabling the creation of autonomous systems that redefine the boundaries of innovation. RISC-V is central to this transformation, offering open and efficient architectures optimised for cutting-edge AI applications.
The Edge AI Academy Summer School 2025 serves as a dynamic platform connecting Master graduates, PhD students, post-doctoral researchers, and industry practitioners passionate about Edge AI and RISC-V.
Participants can explore new concepts, ideas and methods shaping the future of edge intelligence and computing.
Key topics span a spectrum of exciting developments, from hardware acceleration and heterogeneous system integration to advanced applications such as autonomous driving powered by large language models and trust in distributed learning. The program delves into transformative developments like multimodal large language models, generative AI on edge, and neuromorphic computing at the micro-edge, fostering a comprehensive understanding of these emerging frontiers.
Designed to be an enriching experience, the Edge AI Academy Summer School 2025 offers an in-depth exploration of the future of edge AI and RISC-V technologies, equipping participants with the knowledge to define the next chapter of AI-driven autonomous systems. This is the perfect environment to share insights, network, and contribute to advancing intelligent systems at the edge.
Edge AI Academy Summer School 2025 Topics
The topics of interest include, but are not limited to, the followings:
- Edge AI Technologies
- RISC-V-based edge AI hardware acceleration
- Edge AI heterogeneous systems integration
- Generative edge AI
- Autonomous driving using LLMs
- Edge AI and machine vision
- Generative AI
- Explainable and generative modelling on the edge
- Neural networks on RISC-V multi-core processors
- Edge AI security and privacy frontier
- Federated Learning
- Privacy in federated learning
- Privacy enhancing technologies for edge AI
- Trust in distributed learning
- Verification of edge AI requirements
- Multimodal large language models
- Vision transformers for the edge
- Edge deep neural network development frameworks
- IoT and edge AI
- Hardware-efficient architectures for deep neural networks
- Neuromorphic computing at the micro-edge.
- From edge fixed AI to generative AI
- Benchmarking and profiling of large vision models at the edge
- Deep neural networks in RISC-V architectures
Edge AI and RISC-V technologies future trends and roadmap
Program Committee
- Ovidiu Vermesan, SINTEF, Norway
- Luca Valcarenghi, Sant’Anna School of Advanced Studies, Italy
- Filippo Cugini, CNIT, Italy
- Patrick Pype, NXP Semiconductors, Belgium
- Marcello Coppola, STMicroelectronics, France
- Konstantinos Georgopoulos, Technical University of Crete, Greece
- Iakovos Mavroidis, Technical University of Crete, Greece
- Fabian Chersi, CEA-LIST, France
- Alain Pagani, German Research Center for Artificial Intelligence – DFKI, Germany
- Roy Bahr, SINTEF, Norway
