Physics Inspired Criterion for Pruning-Quantization Joint Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Xie, Weiying, Fan, Xiaoyi, Zhang, Xin, Li, Yunsong, Lei, Jie, Fang, Leyuan
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913375788728320
author Xie, Weiying
Fan, Xiaoyi
Zhang, Xin
Li, Yunsong
Lei, Jie
Fang, Leyuan
author_facet Xie, Weiying
Fan, Xiaoyi
Zhang, Xin
Li, Yunsong
Lei, Jie
Fang, Leyuan
contents Pruning-quantization joint learning always facilitates the deployment of deep neural networks (DNNs) on resource-constrained edge devices. However, most existing methods do not jointly learn a global criterion for pruning and quantization in an interpretable way. In this paper, we propose a novel physics inspired criterion for pruning-quantization joint learning (PIC-PQ), which is explored from an analogy we first draw between elasticity dynamics (ED) and model compression (MC). Specifically, derived from Hooke's law in ED, we establish a linear relationship between the filters' importance distribution and the filter property (FP) by a learnable deformation scale in the physics inspired criterion (PIC). Furthermore, we extend PIC with a relative shift variable for a global view. To ensure feasibility and flexibility, available maximum bitwidth and penalty factor are introduced in quantization bitwidth assignment. Experiments on benchmarks of image classification demonstrate that PIC-PQ yields a good trade-off between accuracy and bit-operations (BOPs) compression ratio e.g., 54.96X BOPs compression ratio in ResNet56 on CIFAR10 with 0.10% accuracy drop and 53.24X in ResNet18 on ImageNet with 0.61% accuracy drop). The code will be available at https://github.com/fanxxxxyi/PIC-PQ.
format Preprint
id arxiv_https___arxiv_org_abs_2312_00851
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Physics Inspired Criterion for Pruning-Quantization Joint Learning
Xie, Weiying
Fan, Xiaoyi
Zhang, Xin
Li, Yunsong
Lei, Jie
Fang, Leyuan
Machine Learning
Computer Vision and Pattern Recognition
Pruning-quantization joint learning always facilitates the deployment of deep neural networks (DNNs) on resource-constrained edge devices. However, most existing methods do not jointly learn a global criterion for pruning and quantization in an interpretable way. In this paper, we propose a novel physics inspired criterion for pruning-quantization joint learning (PIC-PQ), which is explored from an analogy we first draw between elasticity dynamics (ED) and model compression (MC). Specifically, derived from Hooke's law in ED, we establish a linear relationship between the filters' importance distribution and the filter property (FP) by a learnable deformation scale in the physics inspired criterion (PIC). Furthermore, we extend PIC with a relative shift variable for a global view. To ensure feasibility and flexibility, available maximum bitwidth and penalty factor are introduced in quantization bitwidth assignment. Experiments on benchmarks of image classification demonstrate that PIC-PQ yields a good trade-off between accuracy and bit-operations (BOPs) compression ratio e.g., 54.96X BOPs compression ratio in ResNet56 on CIFAR10 with 0.10% accuracy drop and 53.24X in ResNet18 on ImageNet with 0.61% accuracy drop). The code will be available at https://github.com/fanxxxxyi/PIC-PQ.
title Physics Inspired Criterion for Pruning-Quantization Joint Learning
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.00851