کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
11030119 | 1646383 | 2019 | 39 صفحه PDF | دانلود رایگان |
عنوان انگلیسی مقاله ISI
FitCNN: A cloud-assisted and low-cost framework for updating CNNs on IoT devices
دانلود مقاله + سفارش ترجمه
دانلود مقاله ISI انگلیسی
رایگان برای ایرانیان
موضوعات مرتبط
مهندسی و علوم پایه
مهندسی کامپیوتر
نظریه محاسباتی و ریاضیات
پیش نمایش صفحه اول مقاله
چکیده انگلیسی
Therefore, this paper proposes a cloud-assisted CNN framework, named FitCNN, with incremental learning and low data transmission, to reduce the overhead of updating CNNs deployed on devices. To reduce the data transmission during incremental learning, we propose a strategy, called Distiller, to selectively upload the data that is worth learning, and develop an extracting strategy, called Juicer, to choose light amount of weights from the new CNN model generated on the cloud to update the corresponding old ones on devices. Experimental results show that the Distiller strategy can reduce 39.4% data transmission of uploading based on a certain dataset, and the Juicer strategy reduces by more than 60% data transmission of updating with multiple CNNs and datasets.
ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Future Generation Computer Systems - Volume 91, February 2019, Pages 277-289
Journal: Future Generation Computer Systems - Volume 91, February 2019, Pages 277-289
نویسندگان
Duo Liu, Chaoshu Yang, Shiming Li, Xianzhang Chen, Jinting Ren, Renping Liu, Moming Duan, Yujuan Tan, Liang Liang,