FantastIC4: A Hardware-Software Co-Design Approach for Efficiently Running 4Bit-Compact Multilayer Perceptrons

With the growing demand for deploying Deep Learning models to the “edge”, it is paramount to develop techniques that allow to execute state-of-the-art models within very tight and limited resource constraints. In this work we propose a software-hardware optimization paradigm fo...

Full description

Bibliographic Details
Main Authors: Simon Wiedemann, Suhas Shivapakash, Daniel Becking, Pablo Wiedemann, Wojciech Samek, Friedel Gerfers, Thomas Wiegand
Format: Article
Language:English
Published: IEEE 2021-01-01
Series:IEEE Open Journal of Circuits and Systems
Subjects:
Online Access:https://ieeexplore.ieee.org/document/9440253/