國立虎尾科技大學 |

Kernel Approximation Methods for Speech Recognition.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Kernel Approximation Methods for Speech Recognition./
作者:	May, Avner.
面頁冊數:	1 online resource (155 pages)
附註:	Source: Dissertation Abstracts International, Volume: 79-04(E), Section: B.
Contained By:	Dissertation Abstracts International79-04B(E).
標題:	Computer science. -
電子資源:	click for full text (PQDT)
ISBN:	9780355540017

Kernel Approximation Methods for Speech Recognition.
May, Avner.

Kernel Approximation Methods for Speech Recognition. - 1 online resource (155 pages)

Source: Dissertation Abstracts International, Volume: 79-04(E), Section: B.

Thesis (Ph.D.)

Includes bibliographical references

Over the past five years or so, deep learning methods have dramatically improved the state of the art performance in a variety of domains, including speech recognition, computer vision, and natural language processing. Importantly, however, they suffer from a number of drawbacks: 1. Training these models is a non-convex optimization problem, and thus it is difficult to guarantee that a trained model minimizes the desired loss function. 2. These models are difficult to interpret. In particular, it is difficult to explain, for a given model, why the computations it performs make accurate predictions.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355540017Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

Kernel Approximation Methods for Speech Recognition.
LDR:03378ntm a2200373Ki 4500 001 909208
005 20180419121558.5
006 m o u
007 cr mn||||a|a||
008 190606s2018 xx obm 000 0 eng d
020 $a 9780355540017
035 $a (MiAaPQ)AAI10685472
035 $a (MiAaPQ)columbia:14355
035 $a AAI10685472
040 $a MiAaPQ $b eng $c MiAaPQ
099 $a TUL $f hyy $c available through World Wide Web
100 1 $a May, Avner. $3 1179844
245 1 0 $a Kernel Approximation Methods for Speech Recognition.
264 0 $c 2018
300 $a 1 online resource (155 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 79-04(E), Section: B.
500 $a Adviser: Michael J. Collins.
502 $a Thesis (Ph.D.) $c Columbia University $d 2018.
504 $a Includes bibliographical references
520 $a Over the past five years or so, deep learning methods have dramatically improved the state of the art performance in a variety of domains, including speech recognition, computer vision, and natural language processing. Importantly, however, they suffer from a number of drawbacks: 1. Training these models is a non-convex optimization problem, and thus it is difficult to guarantee that a trained model minimizes the desired loss function. 2. These models are difficult to interpret. In particular, it is difficult to explain, for a given model, why the computations it performs make accurate predictions.
520 $a In contrast, kernel methods are straightforward to interpret, and training them is a convex optimization problem. Unfortunately, solving these optimization problems exactly is typically prohibitively expensive, though one can use approximation methods to circumvent this problem. In this thesis, we explore to what extent kernel approximation methods can compete with deep learning, in the context of large-scale prediction tasks. Our contributions are as follows: 1. We perform the most extensive set of experiments to date using kernel approximation methods in the context of large-scale speech recognition tasks, and compare performance with deep neural networks. 2. We propose a feature selection algorithm which significantly improves the performance of the kernel models, making their performance competitive with fully-connected feedforward neural networks. 3. We perform an in-depth comparison between two leading kernel approximation strategies --- random Fourier features [Rahimi and Recht, 2007] and the Nystrom method [Williams and Seeger, 2001] --- showing that although the Nystrom method is better at approximating the kernel, it performs worse than random Fourier features when used for learning.
520 $a We believe this work opens the door for future research to continue to push the boundary of what is possible with kernel methods. This research direction will also shed light on the question of when, if ever, deep models are needed for attaining strong performance.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Artificial intelligence. $3 559380
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Columbia University. $b Computer Science. $3 1179509
773 0 $t Dissertation Abstracts International $g 79-04B(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10685472 $z click for full text (PQDT)