Hyperdrive: A Systolically Scalable Binary-Weight CNN Inference Engine for mW IoT End-Nodes
Authors:
- Renzo AndriRARenzo Andri
- Lukas CavigelliLCLukas Cavigelli
- Davide RossiDRDavide Rossi
- Luca BeniniLBLuca Benini
- 1 July 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 509-515
Abstract
Deep neural networks have achieved impressive results in computer vision and machine learning. Unfortunately, state-of-the-art networks are extremely compute-and memory-intensive which makes them unsuitable for mW-devices such as IoT end-nodes. Aggressive quantization of these networks dramatically reduces the computation and memory footprint. Binary-weight neural networks (BWNs) follow this trend, pushing weight quantization to the limit. Hardware accelerators for BWNs presented up to now have focused on core efficiency, disregarding I/O bandwidth and system-level efficiency that are crucial for deployment of accelerators in ultra-low power devices. We present Hyperdrive: a BWN accelerator dramatically reducing the I/O bandwidth exploiting a novel binary-weight streaming approach, and capable of handling high-resolution images by virtue of its systolic-scalable architecture. We achieve a 5.9 TOp/s/W system-level efficiency (i.e. including I/Os)-2.2x higher than state-of-the-art BNN accelerators, even if our core uses resource-intensive FP16 arithmetic for increased robustness.Keywords
All Related Versions
This publication has 25 references indexed in Scilit:
- Origami: A 803-GOp/s/W Convolutional Network AcceleratorIEEE Transactions on Circuits and Systems for Video Technology, 2016
- EIEACM SIGARCH Computer Architecture News, 2016
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- 14.5 Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks2016
- ShiDianNao2015
- Accelerating real-time embedded scene labeling with convolutional networks2015
- Fully convolutional networks for semantic segmentation2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- More Is Less: Signal Processing and the Data DelugeScience, 2011
- Near-Threshold Computing: Reclaiming Moore's Law Through Energy Efficient Integrated CircuitsProceedings of the IEEE, 2010