國立高雄大學圖資館 |

登入

回首頁

Machine Learning in IoT Systems: From Deep Learning to Hyperdimensional Computing.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Machine Learning in IoT Systems: From Deep Learning to Hyperdimensional Computing.
作者:	Imani, Mohsen.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, 2020
面頁冊數:	145 p.
附註:	Source: Dissertations Abstracts International, Volume: 82-02, Section: B.
附註:	Advisor: Rosing, Tajana Simunic.
Contained By:	Dissertations Abstracts International82-02B.
標題:	Computer engineering.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27958350
ISBN:	9798662473379

Machine Learning in IoT Systems: From Deep Learning to Hyperdimensional Computing.
Imani, Mohsen.

Machine Learning in IoT Systems: From Deep Learning to Hyperdimensional Computing. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 145 p.

Source: Dissertations Abstracts International, Volume: 82-02, Section: B.

Thesis (Ph.D.)--University of California, San Diego, 2020.

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

With the emergence of the Internet of Things (IoT), devices are generating massive amounts of data. Running machine learning algorithms on IoT devices poses substantial technical challenges due to their limited resources. The focus of this dissertation is to dramatically increase computing efﬁciency as well as the learning capability of today’s IoT systems by accelerating existing algorithms in hardware and designing new classes of light-weight machine learning algorithms. Our design makes a modiﬁcation to storage-class memory to support search-based and vector-based computation in memory. We show how this architecture can be used to accelerate deep neural networks in both training and inference phases, resulting in 303x faster and 48x more energy efﬁcient training as compared to the state-of-the-art GPU.Hardware acceleration alone does not provide all the efﬁciency and robustness that we need. Therefore, we present Hyperdimensional (HD) computing, an alternative method of learning that implements principles of the functionality in the brain: (i) fast learning, (ii) robustness to noise/error, and (iii) intertwined memory and logic. These features make HD computing a promising solution for today’s embedded devices with limited resources as well as future computing systems in deep nanoscaled technology that have issues of high noise and variability. We exploit emerging technologies to enable processing in-memory which is capable of highly-parallel computation and data movement reduction. Our evaluations show that HD computing provides 39x faster and 56x more energy efﬁciency as compared to state-of-the-art deep learning accelerator.

ISBN: 9798662473379Subjects--Topical Terms:

212944
Computer engineering.
Subjects--Index Terms:

Brain-inspired computing

Machine Learning in IoT Systems: From Deep Learning to Hyperdimensional Computing.
LDR:02935nmm a2200397 4500 001 616367
005 20220513114318.5
008 220920s2020 ||||||||||||||||| ||eng d
020 $a 9798662473379
035 $a (MiAaPQ)AAI27958350
035 $a AAI27958350
040 $a MiAaPQ $c MiAaPQ
100 1 $a Imani, Mohsen. $3 915520
245 1 0 $a Machine Learning in IoT Systems: From Deep Learning to Hyperdimensional Computing.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 145 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-02, Section: B.
500 $a Advisor: Rosing, Tajana Simunic.
502 $a Thesis (Ph.D.)--University of California, San Diego, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a With the emergence of the Internet of Things (IoT), devices are generating massive amounts of data. Running machine learning algorithms on IoT devices poses substantial technical challenges due to their limited resources. The focus of this dissertation is to dramatically increase computing efﬁciency as well as the learning capability of today’s IoT systems by accelerating existing algorithms in hardware and designing new classes of light-weight machine learning algorithms. Our design makes a modiﬁcation to storage-class memory to support search-based and vector-based computation in memory. We show how this architecture can be used to accelerate deep neural networks in both training and inference phases, resulting in 303x faster and 48x more energy efﬁcient training as compared to the state-of-the-art GPU.Hardware acceleration alone does not provide all the efﬁciency and robustness that we need. Therefore, we present Hyperdimensional (HD) computing, an alternative method of learning that implements principles of the functionality in the brain: (i) fast learning, (ii) robustness to noise/error, and (iii) intertwined memory and logic. These features make HD computing a promising solution for today’s embedded devices with limited resources as well as future computing systems in deep nanoscaled technology that have issues of high noise and variability. We exploit emerging technologies to enable processing in-memory which is capable of highly-parallel computation and data movement reduction. Our evaluations show that HD computing provides 39x faster and 56x more energy efﬁciency as compared to state-of-the-art deep learning accelerator.
590 $a School code: 0033.
650 4 $a Computer engineering. $3 212944
650 4 $a Computer science. $3 199325
650 4 $a Electrical engineering. $3 454503
650 4 $a Artificial intelligence. $3 194058
653 $a Brain-inspired computing
653 $a Computer architecture
653 $a Energy efficiency
653 $a Internet of Things
653 $a Machine learning
653 $a Processing in-memory
690 $a 0464
690 $a 0984
690 $a 0544
690 $a 0800
710 2 $a University of California, San Diego. $b Computer Science and Engineering. $3 857483
773 0 $t Dissertations Abstracts International $g 82-02B.
790 $a 0033
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27958350