Anzeige
Erschienen in:
2020 | OriginalPaper | Buchkapitel
Implementing Machine Learning on Edge Devices with Limited Working Memory
verfasst von : A. Harish, Saksham Jhawar, B. S. Anisha, P. Ramakanth Kumar
Erschienen in: Inventive Communication and Computational Technologies
Verlag: Springer Singapore
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
The architecture is aimed at pushing the computing towards the edge. When most of the computation occurs towards the edge device where the data is generated, the processing becomes faster and more efficient. This improves the user’s wait time and delivers results faster to the user. Machine learning techniques are implemented in the edge devices. While one can process data at the sensor, what one can do is limited by the processing power available on each IoT device. Data is at the heart of an IoT architecture, and one needs to choose between immediacy and depth of insight when processing that data. The more immediate the need for information, the closer to the end devices your processing needs to be. We propose an architecture to use machine learning algorithms in the limited memory of the edge device.