Quality at the Tail of Machine Learning Inference

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Zhengxin, Gao, Wanling, Luo, Chunjie, Wang, Lei, Tang, Fei, Wen, Xu, Zhan, Jianfeng
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916136335966208
author Yang, Zhengxin
Gao, Wanling
Luo, Chunjie
Wang, Lei
Tang, Fei
Wen, Xu
Zhan, Jianfeng
author_facet Yang, Zhengxin
Gao, Wanling
Luo, Chunjie
Wang, Lei
Tang, Fei
Wen, Xu
Zhan, Jianfeng
contents Machine learning inference should be subject to stringent inference time constraints while ensuring high inference quality, especially in safety-critical (e.g., autonomous driving) and mission-critical (e.g., emotion recognition) contexts. Neglecting either aspect can lead to severe consequences, such as loss of life and property damage. Many studies lack a comprehensive consideration of these metrics, leading to incomplete or misleading evaluations. The study unveils a counterintuitive revelation: deep learning inference quality exhibits fluctuations due to inference time. To depict this phenomenon, the authors coin a new term, "tail quality," providing a more comprehensive evaluation, and overcoming conventional metric limitations. Moreover, the research proposes an initial evaluation framework to analyze factors affecting quality fluctuations, facilitating the prediction of the potential distribution of inference quality. The effectiveness of the evaluation framework is validated through experiments conducted on deep learning models for three different tasks across four systems.
format Preprint
id arxiv_https___arxiv_org_abs_2212_13925
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Quality at the Tail of Machine Learning Inference
Yang, Zhengxin
Gao, Wanling
Luo, Chunjie
Wang, Lei
Tang, Fei
Wen, Xu
Zhan, Jianfeng
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Software Engineering
Machine learning inference should be subject to stringent inference time constraints while ensuring high inference quality, especially in safety-critical (e.g., autonomous driving) and mission-critical (e.g., emotion recognition) contexts. Neglecting either aspect can lead to severe consequences, such as loss of life and property damage. Many studies lack a comprehensive consideration of these metrics, leading to incomplete or misleading evaluations. The study unveils a counterintuitive revelation: deep learning inference quality exhibits fluctuations due to inference time. To depict this phenomenon, the authors coin a new term, "tail quality," providing a more comprehensive evaluation, and overcoming conventional metric limitations. Moreover, the research proposes an initial evaluation framework to analyze factors affecting quality fluctuations, facilitating the prediction of the potential distribution of inference quality. The effectiveness of the evaluation framework is validated through experiments conducted on deep learning models for three different tasks across four systems.
title Quality at the Tail of Machine Learning Inference
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Software Engineering
url https://arxiv.org/abs/2212.13925