{"title":"OptDNN: Automatic deep neural networks optimizer for edge computing","authors":"Luca Giovannesi, Gabriele Proietti Mattia, Roberto Beraldi","doi":"10.1016/j.simpa.2024.100641","DOIUrl":null,"url":null,"abstract":"<div><p>DNNs are widely used for complex tasks like image and signal processing, and they are in increasing demand for implementation on Internet of Things (IoT) devices. For these devices, optimizing DNN models is a necessary task. Generally, standard optimization approaches require specialists to manually fine-tune hyper-parameters to find a good trade-off between efficiency and accuracy. In this paper, we propose OptDNN, a software that employs innovative and automatic approaches to determine optimal hyper-parameters for pruning, clustering, and quantization. The models optimized by OptDNN have a smaller memory footprint, faster inference time, and a similar accuracy to the original models.</p></div>","PeriodicalId":29771,"journal":{"name":"Software Impacts","volume":"20 ","pages":"Article 100641"},"PeriodicalIF":1.3000,"publicationDate":"2024-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2665963824000290/pdfft?md5=9408edc33cd6715a12afa1a8f06365fc&pid=1-s2.0-S2665963824000290-main.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Software Impacts","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2665963824000290","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, SOFTWARE ENGINEERING","Score":null,"Total":0}
引用次数: 0
Abstract
DNNs are widely used for complex tasks like image and signal processing, and they are in increasing demand for implementation on Internet of Things (IoT) devices. For these devices, optimizing DNN models is a necessary task. Generally, standard optimization approaches require specialists to manually fine-tune hyper-parameters to find a good trade-off between efficiency and accuracy. In this paper, we propose OptDNN, a software that employs innovative and automatic approaches to determine optimal hyper-parameters for pruning, clustering, and quantization. The models optimized by OptDNN have a smaller memory footprint, faster inference time, and a similar accuracy to the original models.