Leveraging Temporal Patterns: Automated Augmentation to Create Temporal Early Exit Networks for Efficient Edge AI

M Sponner, L Servadei, B Waschneck, R Wille… - IEEE …, 2024 - ieeexplore.ieee.org
In embedded systems, efficient deep learning solutions are crucial to balance accuracy and
resource constraints. Early Exit Neural Networks offer a promising solution, but their manual …

[PDF][PDF] PEAX-A Model Augmentation Framework for Adaptive Techniques and Embedded Applications

M Sponner, L Servadei, B Waschneck, R Wille… - 2025 - cfaed.tu-dresden.de
Deploying deep learning models to the edge is challenging due to limited available
resources of embedded devices. To address this, we introduce PEAX, a flexible and …