skip to content
View online Improving neural networks by preventing co-adaptation... Preview this item
ClosePreview this item
Checking...

Improving neural networks by preventing co-adaptation of feature detectors

Author: Hinton, Geoffrey E.; Srivastava, Nitish; Krizhevsky, Alex; Sutskever, Ilya; Salakhutdinov, Ruslan R.
Publisher: 2012-07-03
Edition/Format:   Downloadable archival material
Summary:
When a large feedforward neural network is trained on a small training set, it typically performs poorly on held-out test data. This "overfitting" is greatly reduced by randomly omitting half of the feature detectors on each training case. This prevents complex co-adaptations in which a feature detector is only helpful in the context of several other specific feature detectors. Instead, each neuron learns to detect  Read more...
Rating:

(not yet rated) 0 with reviews - Be the first.

Subjects
More like this

Find a copy online

Links to this item

Find a copy in the library

&AllPage.SpinnerRetrieving; Finding libraries that hold this item...

Details

Genre/Form: text
Material Type: Internet resource
Document Type: Internet Resource, Archival Material
All Authors / Contributors: Hinton, Geoffrey E.; Srivastava, Nitish; Krizhevsky, Alex; Sutskever, Ilya; Salakhutdinov, Ruslan R.
OCLC Number: 798541308

Abstract:

When a large feedforward neural network is trained on a small training set, it typically performs poorly on held-out test data. This "overfitting" is greatly reduced by randomly omitting half of the feature detectors on each training case. This prevents complex co-adaptations in which a feature detector is only helpful in the context of several other specific feature detectors. Instead, each neuron learns to detect a feature that is generally helpful for producing the correct answer given the combinatorially large variety of internal contexts in which it must operate. Random "dropout" gives big improvements on many benchmark tasks and sets new records for speech and object recognition.

Reviews

User-contributed reviews
Retrieving GoodReads reviews...
Retrieving DOGObooks reviews...

Tags

Be the first.
Confirm this request

You may have already requested this item. Please select Ok if you would like to proceed with this request anyway.

Linked Data


Primary Entity

<http://www.worldcat.org/oclc/798541308> # Improving neural networks by preventing co-adaptation of feature detectors
    a schema:MediaObject, schema:CreativeWork, library:ArchiveMaterial ;
   library:oclcnum "798541308" ;
   schema:about <http://experiment.worldcat.org/entity/work/data/1124006828#Thing/computer_science_computer_vision_and_pattern_recognition> ; # Computer Science - Computer Vision and Pattern Recognition
   schema:about <http://experiment.worldcat.org/entity/work/data/1124006828#Thing/computer_science_neural_and_evolutionary_computing> ; # Computer Science - Neural and Evolutionary Computing
   schema:about <http://experiment.worldcat.org/entity/work/data/1124006828#Thing/computer_science_learning> ; # Computer Science - Learning
   schema:creator <http://experiment.worldcat.org/entity/work/data/1124006828#Agent/krizhevsky_alex> ; # Krizhevsky, Alex
   schema:creator <http://experiment.worldcat.org/entity/work/data/1124006828#Agent/salakhutdinov_ruslan_r> ; # Salakhutdinov, Ruslan R.
   schema:creator <http://experiment.worldcat.org/entity/work/data/1124006828#Agent/sutskever_ilya> ; # Sutskever, Ilya
   schema:creator <http://experiment.worldcat.org/entity/work/data/1124006828#Agent/srivastava_nitish> ; # Srivastava, Nitish
   schema:creator <http://experiment.worldcat.org/entity/work/data/1124006828#Agent/hinton_geoffrey_e> ; # Hinton, Geoffrey E.
   schema:datePublished "2012" ;
   schema:datePublished "2012/07/03" ;
   schema:description "When a large feedforward neural network is trained on a small training set, it typically performs poorly on held-out test data. This "overfitting" is greatly reduced by randomly omitting half of the feature detectors on each training case. This prevents complex co-adaptations in which a feature detector is only helpful in the context of several other specific feature detectors. Instead, each neuron learns to detect a feature that is generally helpful for producing the correct answer given the combinatorially large variety of internal contexts in which it must operate. Random "dropout" gives big improvements on many benchmark tasks and sets new records for speech and object recognition." ;
   schema:exampleOfWork <http://worldcat.org/entity/work/id/1124006828> ;
   schema:genre "text" ;
   schema:name "Improving neural networks by preventing co-adaptation of feature detectors" ;
   schema:productID "798541308" ;
   schema:publication <http://www.worldcat.org/title/-/oclc/798541308#PublicationEvent/2012_07_03> ;
   schema:url <http://arxiv.org/abs/1207.0580> ;
   wdrs:describedby <http://www.worldcat.org/title/-/oclc/798541308> ;
    .


Related Entities

<http://experiment.worldcat.org/entity/work/data/1124006828#Agent/hinton_geoffrey_e> # Hinton, Geoffrey E.
    a bgn:Agent ;
   schema:name "Hinton, Geoffrey E." ;
    .

<http://experiment.worldcat.org/entity/work/data/1124006828#Agent/krizhevsky_alex> # Krizhevsky, Alex
    a bgn:Agent ;
   schema:name "Krizhevsky, Alex" ;
    .

<http://experiment.worldcat.org/entity/work/data/1124006828#Agent/salakhutdinov_ruslan_r> # Salakhutdinov, Ruslan R.
    a bgn:Agent ;
   schema:name "Salakhutdinov, Ruslan R." ;
    .

<http://experiment.worldcat.org/entity/work/data/1124006828#Agent/srivastava_nitish> # Srivastava, Nitish
    a bgn:Agent ;
   schema:name "Srivastava, Nitish" ;
    .

<http://experiment.worldcat.org/entity/work/data/1124006828#Agent/sutskever_ilya> # Sutskever, Ilya
    a bgn:Agent ;
   schema:name "Sutskever, Ilya" ;
    .

<http://experiment.worldcat.org/entity/work/data/1124006828#Thing/computer_science_computer_vision_and_pattern_recognition> # Computer Science - Computer Vision and Pattern Recognition
    a schema:Thing ;
   schema:name "Computer Science - Computer Vision and Pattern Recognition" ;
    .

<http://experiment.worldcat.org/entity/work/data/1124006828#Thing/computer_science_learning> # Computer Science - Learning
    a schema:Thing ;
   schema:name "Computer Science - Learning" ;
    .

<http://experiment.worldcat.org/entity/work/data/1124006828#Thing/computer_science_neural_and_evolutionary_computing> # Computer Science - Neural and Evolutionary Computing
    a schema:Thing ;
   schema:name "Computer Science - Neural and Evolutionary Computing" ;
    .


Content-negotiable representations

Close Window

Please sign in to WorldCat 

Don't have an account? You can easily create a free account.