國立虎尾科技大學 |

Transformation Invariance and Equivariance in Deep Learning.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Transformation Invariance and Equivariance in Deep Learning./
作者:	Shen, Jiajun.
面頁冊數:	1 online resource (120 pages)
附註:	Source: Dissertation Abstracts International, Volume: 79-03(E), Section: B.
Contained By:	Dissertation Abstracts International79-03B(E).
標題:	Computer science. -
電子資源:	click for full text (PQDT)
ISBN:	9780355519754

Transformation Invariance and Equivariance in Deep Learning.
Shen, Jiajun.

Transformation Invariance and Equivariance in Deep Learning. - 1 online resource (120 pages)

Source: Dissertation Abstracts International, Volume: 79-03(E), Section: B.

Thesis (Ph.D.)

Includes bibliographical references

A major challenge for object recognition is to correctly perceive the image objects despite the extraneous variations in the data such as shifting, rotation, deformation, etc. It would be much easier for the vision tasks if such task-irrelevant transformation variabilities were removed from the data. The recent success of deep learning approaches has its roots in the ability to build feature representations that are invariant to variations caused by nuisance factors. The expressiveness of deep networks allows the models to disentangle the underlying factors of variations in the data and the training signals guide the models to learn feature representations that are robust to task-irrelevant variations. However, such variations need to be observed from the training data. Otherwise, a traditional deep neural network without special architectural design would not generalize to these variations. To address this concern, we study the problem of achieving transformation invariance and equivariance in deep learning.

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

Mode of access: World Wide Web

ISBN: 9780355519754Subjects--Topical Terms:

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

554714
Electronic books.

Transformation Invariance and Equivariance in Deep Learning.
LDR:03913ntm a2200373Ki 4500 001 910585
005 20180517123957.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
020 $a 9780355519754
035 $a (MiAaPQ)AAI10638757
035 $a (MiAaPQ)uchicago:14115
035 $a AAI10638757
040 $a MiAaPQ $b eng $c MiAaPQ
099 $a TUL $f hyy $c available through World Wide Web
100 1 $a Shen, Jiajun. $3 1181946
245 1 0 $a Transformation Invariance and Equivariance in Deep Learning.
264 0 $c 2017
300 $a 1 online resource (120 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: 79-03(E), Section: B.
500 $a Adviser: Yali Amit.
502 $a Thesis (Ph.D.) $c The University of Chicago $d 2017.
504 $a Includes bibliographical references
520 $a A major challenge for object recognition is to correctly perceive the image objects despite the extraneous variations in the data such as shifting, rotation, deformation, etc. It would be much easier for the vision tasks if such task-irrelevant transformation variabilities were removed from the data. The recent success of deep learning approaches has its roots in the ability to build feature representations that are invariant to variations caused by nuisance factors. The expressiveness of deep networks allows the models to disentangle the underlying factors of variations in the data and the training signals guide the models to learn feature representations that are robust to task-irrelevant variations. However, such variations need to be observed from the training data. Otherwise, a traditional deep neural network without special architectural design would not generalize to these variations. To address this concern, we study the problem of achieving transformation invariance and equivariance in deep learning.
520 $a We show how some existing approaches, such as the stacked statistical model with rotat- able features and the spatial transformer network, are imperfect at learning feature repre- sentations that are invariant or equivariant to transformations. To search for an alternative, we develop a training mechanism to learn transformation-invariant feature representations, where feature maps of canonical images are used as soft targets to guide a deep neural network to produce the same feature representations even when the input images are transformed. As a result, our framework can obtain transformation-invariant feature representations and makes it possible to take advantage of unlabeled data that contains an enormous amount of variations. Additionally, we seek architectural changes to the existing deep learning models and propose a framework for training deep neural networks with optimal instantiations. By introducing latent variables to parametrize the transformations of the data for each class, our approach is able to obtain the optimal instantiations while training for the downstream tasks. Another direction we explore is the use of 3D CAD models to render 2D images as a data augmentation approach. Rendering 2D images from 3D models allows for a more compact way of representing an object class and the models are able to observe more data variations during training. Competitive experimental results are demonstrated by these methods, and our analysis shows that they can be promising directions to achieve transformation invariance and equivariance in deep learning.
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
650 4 $a Statistics. $3 556824
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0800
690 $a 0463
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a The University of Chicago. $b Computer Science. $3 845504
773 0 $t Dissertation Abstracts International $g 79-03B(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10638757 $z click for full text (PQDT)