國立高雄大學圖資館 |

登入

回首頁

Efficient Neural Network Based Systems on Mobile and Cloud Platforms.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Efficient Neural Network Based Systems on Mobile and Cloud Platforms.
作者:	Mao, Jiachen.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, 2020
面頁冊數:	140 p.
附註:	Source: Dissertations Abstracts International, Volume: 82-03, Section: B.
附註:	Advisor: Chen, Yiran;Li, Hai.
Contained By:	Dissertations Abstracts International82-03B.
標題:	Computer engineering.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28022458
ISBN:	9798672145822

Efficient Neural Network Based Systems on Mobile and Cloud Platforms.
Mao, Jiachen.

Efficient Neural Network Based Systems on Mobile and Cloud Platforms. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 140 p.

Source: Dissertations Abstracts International, Volume: 82-03, Section: B.

Thesis (Ph.D.)--Duke University, 2020.

This item must not be sold to any third party vendors.

In recent years, machine learning, especially neural networks arouses unprecedented influence in both academia and industry. The reason lies in the state-of-the-art performance of neural networks on many critical applications such as object detection, translation, and games. However, the deployment of neural network models on resource-constrained devices (e.g. edge devices) is challenged by their heavy memory and computing cost during execution. Many efforts have been done in previous literature for efficient execution of neural networks, including the perspectives of hardware, software, and algorithm.My research focus during my Ph.D. study is mainly on software, and algorithm targeting at mobile platforms. More specifically, we emphasize the system design, system optimization, and model compression of neural networks for better mobile user experience. From the system design perspective, we first propose MoDNN – a local distributed mobile computing system for DNN testing. MoDNN can partition already trained DNN models onto several mobile devices to accelerate DNN computations by alleviating device-level computing cost and memory usage. Two model partition schemes are also designed to minimize non-parallel data delivery time, including both wakeup time and transmission time. Then, we propose AdaLearner – an adaptive local distributed mobile computing system for DNN training. To exploit the potential of our system, we adapt the neural networks training phase to mobile device-wise resources and fiercely decrease the transmission overhead for better system scalability. From the system optimization perspective, we propose MobiEye, a cloud-based video detection system optimized for deployment in real-time mobile applications. MobiEye is based on a state-of-the-art video detection framework called Deep Feature Flow (DFF). MobiEye optimizes DFF by three system-level optimization methods. From the model compression perspective, we propose Tprune, a model analyzing and pruning framework for Transformer. In TPrune, we first proposed Block-wise Structured Sparsity Learning (BSSL) to analyze Transformer model property. Then, based on the characters derived from BSSL, we apply Structured Hoyer Square (SHS) to derive the final compressed models. The realization of the projects during my PhD study could contribute to the current research on efficient neural network execution and thus result in more user-friendly and smart applications on edge devices for more users.

ISBN: 9798672145822Subjects--Topical Terms:

212944
Computer engineering.

Efficient Neural Network Based Systems on Mobile and Cloud Platforms.
LDR:03465nmm a2200301 4500 001 616373
005 20220513114320.5
008 220920s2020 ||||||||||||||||| ||eng d
020 $a 9798672145822
035 $a (MiAaPQ)AAI28022458
035 $a AAI28022458
040 $a MiAaPQ $c MiAaPQ
100 1 $a Mao, Jiachen. $3 915528
245 1 0 $a Efficient Neural Network Based Systems on Mobile and Cloud Platforms.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 140 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-03, Section: B.
500 $a Advisor: Chen, Yiran;Li, Hai.
502 $a Thesis (Ph.D.)--Duke University, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a In recent years, machine learning, especially neural networks arouses unprecedented influence in both academia and industry. The reason lies in the state-of-the-art performance of neural networks on many critical applications such as object detection, translation, and games. However, the deployment of neural network models on resource-constrained devices (e.g. edge devices) is challenged by their heavy memory and computing cost during execution. Many efforts have been done in previous literature for efficient execution of neural networks, including the perspectives of hardware, software, and algorithm.My research focus during my Ph.D. study is mainly on software, and algorithm targeting at mobile platforms. More specifically, we emphasize the system design, system optimization, and model compression of neural networks for better mobile user experience. From the system design perspective, we first propose MoDNN – a local distributed mobile computing system for DNN testing. MoDNN can partition already trained DNN models onto several mobile devices to accelerate DNN computations by alleviating device-level computing cost and memory usage. Two model partition schemes are also designed to minimize non-parallel data delivery time, including both wakeup time and transmission time. Then, we propose AdaLearner – an adaptive local distributed mobile computing system for DNN training. To exploit the potential of our system, we adapt the neural networks training phase to mobile device-wise resources and fiercely decrease the transmission overhead for better system scalability. From the system optimization perspective, we propose MobiEye, a cloud-based video detection system optimized for deployment in real-time mobile applications. MobiEye is based on a state-of-the-art video detection framework called Deep Feature Flow (DFF). MobiEye optimizes DFF by three system-level optimization methods. From the model compression perspective, we propose Tprune, a model analyzing and pruning framework for Transformer. In TPrune, we first proposed Block-wise Structured Sparsity Learning (BSSL) to analyze Transformer model property. Then, based on the characters derived from BSSL, we apply Structured Hoyer Square (SHS) to derive the final compressed models. The realization of the projects during my PhD study could contribute to the current research on efficient neural network execution and thus result in more user-friendly and smart applications on edge devices for more users.
590 $a School code: 0066.
650 4 $a Computer engineering. $3 212944
650 4 $a Web studies. $3 708690
690 $a 0464
690 $a 0646
710 2 $a Duke University. $b Electrical and Computer Engineering. $3 531041
773 0 $t Dissertations Abstracts International $g 82-03B.
790 $a 0066
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28022458