國立高雄大學圖資館 |

語系: 繁體中文

說明(常見問題)

圖資館首頁

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Ristretto: Hardware-Oriented Approxi...

Gysel, Philipp Matthias.

Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks.
作者:	Gysel, Philipp Matthias.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, 2016
面頁冊數:	73 p.
附註:	Source: Masters Abstracts International, Volume: 56-02.
附註:	Adviser: Soheil Ghiasi.
Contained By:	Masters Abstracts International56-02(E).
標題:	Electrical engineering.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10165798
ISBN:	9781369201741

Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks.
Gysel, Philipp Matthias.

Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks. - Ann Arbor : ProQuest Dissertations & Theses, 2016 - 73 p.

Source: Masters Abstracts International, Volume: 56-02.

Thesis (M.S.)--University of California, Davis, 2016.

Convolutional neural networks (CNN) have achieved major breakthroughs in recent years. Their performance in computer vision have matched and in some areas even surpassed human capabilities. Deep neural networks can capture complex non-linear features; however this ability comes at the cost of high computational and memory requirements. State-of-art networks require billions of arithmetic operations and millions of parameters.

ISBN: 9781369201741Subjects--Topical Terms:

454503
Electrical engineering.

Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks.
LDR:02797nmm a2200337 4500 001 502154
005 20170619070732.5
008 170818s2016 ||||||||||||||||| ||eng d
020 $a 9781369201741
035 $a (MiAaPQ)AAI10165798
035 $a AAI10165798
040 $a MiAaPQ $c MiAaPQ
100 1 $a Gysel, Philipp Matthias. $3 766204
245 1 0 $a Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks.
260 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2016
300 $a 73 p.
500 $a Source: Masters Abstracts International, Volume: 56-02.
500 $a Adviser: Soheil Ghiasi.
502 $a Thesis (M.S.)--University of California, Davis, 2016.
520 $a Convolutional neural networks (CNN) have achieved major breakthroughs in recent years. Their performance in computer vision have matched and in some areas even surpassed human capabilities. Deep neural networks can capture complex non-linear features; however this ability comes at the cost of high computational and memory requirements. State-of-art networks require billions of arithmetic operations and millions of parameters.
520 $a To enable embedded devices such as smartphones, Google glasses and monitoring cameras with the astonishing power of deep learning, dedicated hardware accelerators can be used to decrease both execution time and power consumption. In applications where fast connection to the cloud is not guaranteed or where privacy is important, computation needs to be done locally. Many hardware accelerators for deep neural networks have been proposed recently. A first important step of accelerator design is hardware-oriented approximation of deep networks, which enables energy-efficient inference.
520 $a We present Ristretto, a fast and automated framework for CNN approximation. Ristretto simulates the hardware arithmetic of a custom hardware accelerator. The framework reduces the bit-width of network parameters and outputs of resource-intense layers, which reduces the chip area for multiplication units significantly. Alternatively, Ristretto can remove the need for multipliers altogether, resulting in an adder-only arithmetic. The tool fine-tunes trimmed networks to achieve high classification accuracy.
520 $a Since training of deep neural networks can be time-consuming, Ristretto uses highly optimized routines which run on the GPU. This enables fast compression of any given network.
520 $a Given a maximum tolerance of 1%, Ristretto can successfully condense CaffeNet and SqueezeNet to 8-bit. The code for Ristretto is available.
590 $a School code: 0029.
650 4 $a Electrical engineering. $3 454503
650 4 $a Computer engineering. $3 212944
690 $a 0544
690 $a 0464
710 2 $a University of California, Davis. $b Electrical and Computer Engineering. $3 531094
773 0 $t Masters Abstracts International $g 56-02(E).
790 $a 0029
791 $a M.S.
792 $a 2016
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10165798