ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Journal Article Quantune: Post-training Quantization of Convolutional Neural Networks using Extreme Gradient Boosting for Fast Deployment
Cited 10 time in scopus Download 2283 time Share share facebook twitter linkedin kakaostory
Authors
Lee Jemin, Yu Misun, Kwon Yongin, Taeho Kim
Issue Date
202207
Source
Future Generation Computer Systems, v.132, pp.124-135
ISSN
0167-739X
Publisher
Elsevier
DOI
https://dx.doi.org/10.1016/j.future.2022.02.005
Project Code
21HS4700, Neuromorphic Computing Software Platform for Artificial Intelligence Systems, Taeho Kim
Abstract
To adopt convolutional neural networks (CNN) for a range of resource-constrained targets, it is necessary to compress the CNN models by performing quantization, whereby precision representation is converted to a lower bit representation. To overcome problems such as sensitivity of the training dataset, high computational requirements, and large time consumption, post-training quantization methods that do not require retraining have been proposed. In addition, to compensate for the accuracy drop without retraining, previous studies on post-training quantization have proposed several complementary methods: calibration, schemes, clipping, granularity, and mixed-precision. To generate a quantized model with minimal error, it is necessary to study all possible combinations of the methods because each of them is complementary and the CNN models have different characteristics. However, an exhaustive or a heuristic search is either too time-consuming or suboptimal. To overcome this challenge, we propose an auto-tuner known as Quantune, which builds a gradient tree boosting model to accelerate the search for the configurations of quantization and reduce the quantization error. We evaluate and compare Quantune with the random, grid, and genetic algorithms. The experimental results show that Quantune reduces the search time for quantization by approximately 36.5× with an accuracy loss of 0.07??0.65% across six CNN models, including the fragile ones (MobileNet, SqueezeNet, and ShuffleNet). To support multiple targets and adopt continuously evolving quantization works, Quantune is implemented on a full-fledged compiler for deep learning as an open-sourced project.
KSP Keywords
Accuracy loss, Bit representation, Complementary methods, Computational requirements, Convolution neural network(CNN), Genetic Algorithm, Quantization error, Resource-constrained, Search time, Time consumption, deep learning(DL)
This work is distributed under the term of Creative Commons License (CCL)
(CC BY NC ND)
CC BY NC ND