Multi-Partner Project
Efficient Deep Learning Platforms for Next-Generation Embedded Edge-AI Systems
Rajendra Bishnoi (TU Delft - Electrical Engineering, Mathematics and Computer Science)
Mohammad Amin Yaldagard (TU Delft - Electrical Engineering, Mathematics and Computer Science)
Konstantinos Stavrakakis (TU Delft - Electrical Engineering, Mathematics and Computer Science)
Said Hamdioui (TU Delft - Electrical Engineering, Mathematics and Computer Science)
Kanishkan Vadivel (IMEC Nederland)
Pankaj Upadhyay (IMEC Nederland)
Nicolas Daniel Rodriguez (Silicon Austria Labs GmbH)
Teresa Van Dam (Almende B.V)
Sander Steeghs-Turchina (Almende B.V)
More Authors
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
The objective of our collaborative multi-partner project is to create an open-source Deep Learning framework called AIDGE for edge and embedded Artificial Intelligence (AI), built around an established European value chain. The framework is designed to support diverse application domains that function independently while serving a broad international community. It offers an integrated, full-stack workflow from Neural Network design and optimization to AI application development and hardware-level implementation with automated code generation for specific hardware targets. The platform aims to provide researchers and developers with a flexible environment to explore novel AI concepts, rapidly prototype solutions, and ensure strong alignment between academic research and industrial requirements. This paper summarizes the progress, outcomes, and milestones achieved up to the second year of this three-year project.