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...
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/ |
Similar Items
-
A Power Efficiency Enhancements of a Multi-Bit Accelerator for Memory Prohibitive Deep Neural Networks
by: Suhas Shivapakash, et al.
Published: (2021-01-01) -
ZEM: Zero-Cycle Bit-Masking Module for Deep Learning Refresh-Less DRAM
by: Duy-Thanh Nguyen, et al.
Published: (2021-01-01) -
Learning Ratio Mask with Cascaded Deep Neural Networks for Echo Cancellation in Laser Monitoring Signals
by: Haitao Lang, et al.
Published: (2020-05-01) -
Design Space Exploration of DNNs for Autonomous Systems
by: Duggal, Jayan Kant
Published: (2019) -
An Energy-Efficient Fall Detection Method Based on FD-DNN for Elderly People
by: Leyuan Liu, et al.
Published: (2020-07-01)