Publication Details
ApproxDARTS: Differentiable Neural Architecture Search with Approximate Multipliers
Sekanina Lukáš, prof. Ing., Ph.D. (DCSY)
Mrázek Vojtěch, Ing., Ph.D. (DCSY)
Neural Architecture Search, Convolutional Neural Networks, Approximate Computing,
Machine Learning
Integrating the principles of approximate computing into the design of
hardware-aware deep neural networks (DNN) has led to DNNs implementations showing
good output quality and highly optimized hardware parameters such as low latency
or inference energy. In this work, we present ApproxDARTS, a neural architecture
search (NAS) method enabling the popular differentiable neural architecture
search method called DARTS to exploit approximate multipliers and thus reduce the
power consumption of generated neural networks.
We showed on the CIFAR-10 data set that the ApproxDARTS is able to perform
a complete architecture search within less than 10 GPU hours and produce
competitive convolutional neural networks (CNN) containing approximate
multipliers in convolutional layers. For example, ApproxDARTS created a CNN
showing an energy consumption reduction of (a) 53.84% in the arithmetic
operations of the inference phase compared to the CNN utilizing the native 32-bit
floating-point multipliers and (b) 5.97% compared to the CNN utilizing the exact
8-bit fixed-point multipliers, in both cases with a negligible accuracy drop.
Moreover, the ApproxDARTS is 2.3 times faster than a similar but evolutionary
algorithm-based method called EvoApproxNAS.
@inproceedings{BUT188465,
author="Michal {Piňos} and Lukáš {Sekanina} and Vojtěch {Mrázek}",
title="ApproxDARTS: Differentiable Neural Architecture Search with Approximate Multipliers",
booktitle="2024 The International Joint Conference on Neural Networks (IJCNN)",
year="2024",
pages="1--8",
publisher="Institute of Electrical and Electronics Engineers",
address="Yokohama",
doi="10.1109/IJCNN60899.2024.10650823",
isbn="979-8-3503-5931-2",
url="https://ieeexplore.ieee.org/document/10650823"
}