Optimising Hardware Accelerated Neural Networks with Quantisation and a Knowledge Distillation Evolutionary Algorithm

Optimising Hardware Accelerated Neural Networks with Quantisation and a Knowledge Distillation Evolutionary Algorithm

This paper compares the latency, accuracy, training time and hardware costs of neural networks compressed with our new multi-objective evolutionary algorithm called NEMOKD, and with quantisation. We evaluate NEMOKD on Intel’s Movidius Myriad X VPU processor, and quantisation on Xilinx’s programmable...

Full description

Bibliographic Details
Main Authors: Robert Stewart, Andrew Nowlan, Pascal Bacchus, Quentin Ducasse, Ekaterina Komendantskaya
Format: Article
Language:English
Published: MDPI AG 2021-02-01
Series:Electronics
Subjects:
quantisation
evolutionary algorithm
neural network
FPGA
Movidius VPU
Online Access:https://www.mdpi.com/2079-9292/10/4/396

Internet

https://www.mdpi.com/2079-9292/10/4/396

Similar Items