Fribourg Linux Seminar

Join us at the 29th Fribourg Linux Seminar, where we explore AI in Embedded Systems.

Our expert speakers will share their insights and experiences, practical advice, and best practices for making use of AI in Embedded Systems so as to leverage their terrific power.

Don't miss this opportunity to network with fellow developers and computer science professionals and learn from the best in the business.

Context
As intelligent technologies become increasingly integrated into the world around us, the fusion of Artificial Intelligence and Embedded Systems is reshaping how we interact with devices—from everyday appliances to critical infrastructure. This conference offers a platform to explore how AI is transforming Embedded Systems into smarter, more adaptive, and context-aware technologies.

We invite you to join us as we examine the challenges, innovations, and real-world applications driving this exciting convergence of AI and Embedded Computing.

Agenda

- 17h00 : Reception of Attendees

- 17h15 : Welcome and Information HEIA-FR (Jacques Supcik, Luca Haab HEIA-FR, Roland Scherwey, Professors HEIA-FR)

- 17h25 : Guessing Better with Less: VAEs in Tiny Places?, Beat Wolf, Professor at HEIA-FR

Modern AI in embedded systems isn’t just pattern recognition—it’s about learning compressed, meaningful representations that still know what’s going on. Variational Autoencoders (VAEs) shrink high-dimensional data into useful embeddings, enabling smart classification, reconstruction, and predictive maintenance. We'll explore how this helps in classification, reconstruction, and even predicting failures before they happen. Because sometimes, guessing well is better than storing everything.

- 17h45 : Inside EYE+: Image Classification for Industrial Applications, Cyril Jacquier - Senior R&D Engineer, Asyril SA

Released in 2021, the EYE+ smart control system quickly gained traction among customers thanks to its remarkably intuitive interface. One key to its success was the ability to easily and efficiently classify parts using AI directly on the device.
This talk explores how this feature was conceived, implemented, and optimized to run reliably on an embedded system. You’ll learn how a carefully balanced combination of software and hardware technologies enabled Asyril SA to deliver a fast, robust, and production-proven solution that continues to power industrial automation today.

- 18h20 : Bring Machine Learning into the Abyss of FPGAs with Vitis AI, Julien Michel - MSE Student and Passionate Geek

In recent years, alongside the explosive rise of Large Language Models (LLMs) in the AI landscape, the embedded industry has followed its own trend going toward bringing MLcloser to the edge and introducing new challenges such as balancing performance and energyefficiency. In this context, FPGAs offer an interesting solution combining parallelism, reconfigurability and relatively low power consumption.
Yet, how can we bring such complex, high-level ML algorithms into the “abyssal” world of hardware design languages (HDL)? And what are the limitations of this integration?
In this talk, we will attempt to answer these questions through a hands-on exploration of AMD’s Vitis AI framework. We will walk through its workflow to highlight the advantages and limitations and briefly examine alternative solutions to this approach.

- 18h50 : Teaser: a message from space, Jacques Supcik, Professor at HEIA-FR and Geek in Chief

- 19h05 : Apéritif and Networking

***

Slides / Languages
EN: The slides will be in English and the talks will be either in EN, FR, or DE.
FR: Les transparents seront en anglais et les présentations pourront être faites en FR, en EN, ou en DE.
DE: Die Folien werden auf Englisch sein und die Vorträge können entweder auf DE, EN oder FR gehalten werden.