Optimize Your AI at the Edge: Explore Hardware Options with dAIEdge-VLab
Workshop Details
Practical Information
This workshop welcomes both decision-makers and developers with two pathways. For high-level exploration: laptop only required. For technical workflow: Python 3.10 installed and a development environment (such as Visual Studio Code).
Two pathways available: high-level exploration (no prerequisites) or technical hands-on (basic understanding of AI/ML pipelines and Python programming familiarity required). Presentation in English, hands-on support available in both English and French.
About this Workshop
The goal of this workshop is to raise the awareness of SMEs and Machine Learning developers on the challenges of deploying AI models on the edge. A hands-on activity will demonstrate how to use the dAIEdge-VLab (https://vlab.daiedge.eu/) to identify the most suitable edge device that meets the model's latency and resource consumption requirements.
The dAIEdge-VLab is a tool designed to benchmark and evaluate the performance of AI models on various edge targets. It provides a user-friendly interface to configure and run benchmarks, visualize results, and compare different models and target edge devices.
In this workshop, participants will learn how to leverage dAIEdge-VLab to benchmark and compare AI models across multiple edge hardware options. Participants will discover how to optimize ML models and evaluate performance, latency, energy consumption, to determine the best fit for the provided use case. By the end of the session, they will be able to confidently identify the most appropriate and cost-effective edge platform for the proposed AI model.
Workshop Agenda
- Introduction
- Advantages and constraints of deploying on the edge
- Presentation of the dAIEdge-VLab
- Hands-on session
- Q&A and Closing Discussion
Speakers & Organizers
Dr. Nuria Pazos Escudero
Head of Embedded Computing Systems Research Group
HE-Arc Ingénierie
Dr. Nuria Pazos Escudero has been working on optimizing and deploying AI on edge devices since 2017 with successful European projects Bonseyes and BonsAPPs. She's now tackling with her team decentralized edge ML training and benchmarking within the European project dAIEdge (dAIEDGE - A network of excellence for distributed, trustworthy, efficient and scalable AI at the Edge).
Maïck Huguenin-Vuillemin
Senior Researcher, Embedded Computing Systems Research Group
HE-Arc Ingénierie
Maïck Huguenin-Vuillemin has worked since 2023 on optimizing and deploying AI on edge devices. He has actively participated in the BonsAPPs project, and he is currently developing the Virtual Lab within the frame of the dAIEdge Network of Excellence.
Margaux Divernois
Senior Researcher, Interaction Technologies Research Group
HE-Arc Ingénierie
Margaux Divernois has worked since 2016 on deploying AI applications for industrial partners, such as creating a digital twin, analyzing software quality and optimizing industrial processes.