ACTIVATION LAYERS FOR DEEP LEARNING NETWORKS

Tasks such as object classification from image data can take advantage of a deep learning process using convolutional neural networks. These networks can include a convolutional layer followed by an activation layer, or activation unit, among other potential layers. Improved accuracy can be obtained...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Tran, Son Dinh, Manmatha, Raghavan
Format:	Patent
Sprache:	eng
Schlagworte:	CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING HANDLING RECORD CARRIERS PHYSICS PRESENTATION OF DATA RECOGNITION OF DATA RECORD CARRIERS
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

container_end_page
container_issue
container_start_page
container_title
container_volume
creator	Tran, Son Dinh Manmatha, Raghavan
description	Tasks such as object classification from image data can take advantage of a deep learning process using convolutional neural networks. These networks can include a convolutional layer followed by an activation layer, or activation unit, among other potential layers. Improved accuracy can be obtained by using a generalized linear unit (GLU) as an activation unit in such a network, where a GLU is linear for both positive and negative inputs, and is defined by a positive slope, a negative slope, and a bias. These parameters can be learned for each channel or a block of channels, and stacking those types of activation units can further improve accuracy.
format	Patent
fullrecord	<record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_US2018197049A1</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>US2018197049A1</sourcerecordid><originalsourceid>FETCH-epo_espacenet_US2018197049A13</originalsourceid><addsrcrecordid>eNrjZNBxdA7xDHMM8fT3U_BxjHQNClZw8w9ScHF1DVDwcXUM8vP0c1fwcw0J9w_yDuZhYE1LzClO5YXS3AzKbq4hzh66qQX58anFBYnJqXmpJfGhwUYGhhaGluYGJpaOhsbEqQIAOm0mTQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>ACTIVATION LAYERS FOR DEEP LEARNING NETWORKS</title><source>esp@cenet</source><creator>Tran, Son Dinh ; Manmatha, Raghavan</creator><creatorcontrib>Tran, Son Dinh ; Manmatha, Raghavan</creatorcontrib><description>Tasks such as object classification from image data can take advantage of a deep learning process using convolutional neural networks. These networks can include a convolutional layer followed by an activation layer, or activation unit, among other potential layers. Improved accuracy can be obtained by using a generalized linear unit (GLU) as an activation unit in such a network, where a GLU is linear for both positive and negative inputs, and is defined by a positive slope, a negative slope, and a bias. These parameters can be learned for each channel or a block of channels, and stacking those types of activation units can further improve accuracy.</description><language>eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; HANDLING RECORD CARRIERS ; PHYSICS ; PRESENTATION OF DATA ; RECOGNITION OF DATA ; RECORD CARRIERS</subject><creationdate>2018</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20180712&DB=EPODOC&CC=US&NR=2018197049A1$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20180712&DB=EPODOC&CC=US&NR=2018197049A1$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>Tran, Son Dinh</creatorcontrib><creatorcontrib>Manmatha, Raghavan</creatorcontrib><title>ACTIVATION LAYERS FOR DEEP LEARNING NETWORKS</title><description>Tasks such as object classification from image data can take advantage of a deep learning process using convolutional neural networks. These networks can include a convolutional layer followed by an activation layer, or activation unit, among other potential layers. Improved accuracy can be obtained by using a generalized linear unit (GLU) as an activation unit in such a network, where a GLU is linear for both positive and negative inputs, and is defined by a positive slope, a negative slope, and a bias. These parameters can be learned for each channel or a block of channels, and stacking those types of activation units can further improve accuracy.</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>HANDLING RECORD CARRIERS</subject><subject>PHYSICS</subject><subject>PRESENTATION OF DATA</subject><subject>RECOGNITION OF DATA</subject><subject>RECORD CARRIERS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2018</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZNBxdA7xDHMM8fT3U_BxjHQNClZw8w9ScHF1DVDwcXUM8vP0c1fwcw0J9w_yDuZhYE1LzClO5YXS3AzKbq4hzh66qQX58anFBYnJqXmpJfGhwUYGhhaGluYGJpaOhsbEqQIAOm0mTQ</recordid><startdate>20180712</startdate><enddate>20180712</enddate><creator>Tran, Son Dinh</creator><creator>Manmatha, Raghavan</creator><scope>EVB</scope></search><sort><creationdate>20180712</creationdate><title>ACTIVATION LAYERS FOR DEEP LEARNING NETWORKS</title><author>Tran, Son Dinh ; Manmatha, Raghavan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_US2018197049A13</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng</language><creationdate>2018</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>HANDLING RECORD CARRIERS</topic><topic>PHYSICS</topic><topic>PRESENTATION OF DATA</topic><topic>RECOGNITION OF DATA</topic><topic>RECORD CARRIERS</topic><toplevel>online_resources</toplevel><creatorcontrib>Tran, Son Dinh</creatorcontrib><creatorcontrib>Manmatha, Raghavan</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Tran, Son Dinh</au><au>Manmatha, Raghavan</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>ACTIVATION LAYERS FOR DEEP LEARNING NETWORKS</title><date>2018-07-12</date><risdate>2018</risdate><abstract>Tasks such as object classification from image data can take advantage of a deep learning process using convolutional neural networks. These networks can include a convolutional layer followed by an activation layer, or activation unit, among other potential layers. Improved accuracy can be obtained by using a generalized linear unit (GLU) as an activation unit in such a network, where a GLU is linear for both positive and negative inputs, and is defined by a positive slope, a negative slope, and a bias. These parameters can be learned for each channel or a block of channels, and stacking those types of activation units can further improve accuracy.</abstract><oa>free_for_read</oa></addata></record>
fulltext	fulltext_linktorsrc
identifier
ispartof
issn
language	eng
recordid	cdi_epo_espacenet_US2018197049A1
source	esp@cenet
subjects	CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING HANDLING RECORD CARRIERS PHYSICS PRESENTATION OF DATA RECOGNITION OF DATA RECORD CARRIERS
title	ACTIVATION LAYERS FOR DEEP LEARNING NETWORKS
url	https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-03T22%3A45%3A04IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=Tran,%20Son%20Dinh&rft.date=2018-07-12&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EUS2018197049A1%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true