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...
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: | |
Online Access: | https://www.mdpi.com/2079-9292/10/4/396 |
Similar Items
-
Optimisation of dynamic quantisation and control for quantised state feedback control system
by: Than Zaw Soe, et al.
Published: (2020-03-01) -
Quantum Cosmology with Third Quantisation
by: Salvador J. Robles-Pérez
Published: (2021-10-01) -
Hybrid 3D Fractal Coding with Neighbourhood Vector Quantisation
by: Yao Zhen, et al.
Published: (2004-01-01) -
Quantisation‐based target detection for cooperative multiple input multiple output radar and communications with multiple stations
by: Zhen Wang, et al.
Published: (2022-09-01) -
Impact of Quantisal® Oral Fluid Collection Device on Drug Stability
by: Michela Riggio, et al.
Published: (2021-07-01)