Improving the Accuracy and Hardware Efficiency of Neural Networks Using Approximate Multipliers

Improving the Accuracy and Hardware Efficiency of Neural Networks Using Approximate Multipliers

Článek WoS

Improvingthe accuracy of a neural network (NN) usually requires using larger hardwarethat consumes more energy. However, the error tolerance of NNs and theirapplications allow approximate computing techniques to be applied to reduceimplementation costs. Given that multiplication is the most resource-intensiveand power-hungry operation in NNs, more economical approximate multipliers(AMs) can significantly reduce hardware costs. In this article, we show that usingAMs can also improve the NN accuracy by introducing noise. We consider twocategories of AMs: 1) deliberately designed and 2) Cartesian genetic programing(CGP)-based AMs. The exact multipliers in two representative NNs, a multilayerperceptron (MLP) and a convolutional NN (CNN), are replaced with approximatedesigns to evaluate their effect on the classification accuracy of the MixedNational Institute of Standards and Technology (MNIST) and Street View HouseNumbers (SVHN) data sets, respectively. Interestingly, up to 0.63% improvementin the classification accuracy is achieved with reductions of 71.45% and 61.55%in the energy consumption and area, respectively. Finally, the features in anAM are identified that tend to make one design outperform others with respectto NN accuracy. Those features are then used to train a predictor thatindicates how well an AM is likely to work in an NN.

Improvingthe accuracy of a neural network (NN) usually requires using larger hardwarethat consumes more energy. However, the error tolerance of NNs and theirapplications allow approximate computing techniques to be applied to reduceimplementation costs. Given that multiplication is the most resource-intensiveand power-hungry operation in NNs, more economical approximate multipliers(AMs) can significantly reduce hardware costs. In this article, we show that usingAMs can also improve the NN accuracy by introducing noise. We consider twocategories of AMs: 1) deliberately designed and 2) Cartesian genetic programing(CGP)-based AMs. The exact multipliers in two representative NNs, a multilayerperceptron (MLP) and a convolutional NN (CNN), are replaced with approximatedesigns to evaluate their effect on the classification accuracy of the MixedNational Institute of Standards and Technology (MNIST) and Street View HouseNumbers (SVHN) data sets, respectively. Interestingly, up to 0.63% improvementin the classification accuracy is achieved with reductions of 71.45% and 61.55%in the energy consumption and area, respectively. Finally, the features in anAM are identified that tend to make one design outperform others with respectto NN accuracy. Those features are then used to train a predictor thatindicates how well an AM is likely to work in an NN.

approximate multipliers, Cartesian genetic programming, convolutional neural network, multi-layer perceptron, neural networks

approximate multipliers, Cartesian genetic programming, convolutional neural network, multi-layer perceptron, neural networks

ANSARI, M.; MRÁZEK, V.; COCKBURN, B.; SEKANINA, L.; VAŠÍČEK, Z.; HAN, J.

2021

22.01.2020

1063-8210

IEEE Transactions on Very Large Scale Integration (VLSI) Systems

28

2

Spojené státy americké

317

328

12

https://www.fit.vut.cz/research/publication/12066/

08863138