Neural network image classifier
A training engine is described which has a memory arranged to access a neural network image classifier, the neural network image classifier having been trained using a plurality of training images from an input space, the training images being labeled for a plurality of classes. The training engine...
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Patent |
| Sprache: | eng |
| Schlagworte: | |
| 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 | Nori, Aditya Vithal Bastani, Osbert Criminisi, Antonio Lampropoulos, Leonidas Vytiniotis, Dimitrios |
| description | A training engine is described which has a memory arranged to access a neural network image classifier, the neural network image classifier having been trained using a plurality of training images from an input space, the training images being labeled for a plurality of classes. The training engine has an adversarial example generator which computes a plurality of adversarial images by, for each adversarial image, searching a region in the input space around one of the training images, the region being one in which the neural network is linear, to find an image which is incorrectly classified into the plurality of classes by the neural network. The training engine has a processor which further trains the neural network image classifier using at least the adversarial images. |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_US10007866B2</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>US10007866B2</sourcerecordid><originalsourceid>FETCH-epo_espacenet_US10007866B23</originalsourceid><addsrcrecordid>eNrjZJD3Sy0tSsxRyEstKc8vylbIzE1MT1VIzkksLs5My0wt4mFgTUvMKU7lhdLcDIpuriHOHrqpBfnxqcUFicmpQK3xocGGBgYG5hZmZk5GxsSoAQA28CUr</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Neural network image classifier</title><source>esp@cenet</source><creator>Nori, Aditya Vithal ; Bastani, Osbert ; Criminisi, Antonio ; Lampropoulos, Leonidas ; Vytiniotis, Dimitrios</creator><creatorcontrib>Nori, Aditya Vithal ; Bastani, Osbert ; Criminisi, Antonio ; Lampropoulos, Leonidas ; Vytiniotis, Dimitrios</creatorcontrib><description>A training engine is described which has a memory arranged to access a neural network image classifier, the neural network image classifier having been trained using a plurality of training images from an input space, the training images being labeled for a plurality of classes. The training engine has an adversarial example generator which computes a plurality of adversarial images by, for each adversarial image, searching a region in the input space around one of the training images, the region being one in which the neural network is linear, to find an image which is incorrectly classified into the plurality of classes by the neural network. The training engine has a processor which further trains the neural network image classifier using at least the adversarial images.</description><language>eng</language><subject>CALCULATING ; COMPUTING ; COUNTING ; PHYSICS</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=20180626&DB=EPODOC&CC=US&NR=10007866B2$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20180626&DB=EPODOC&CC=US&NR=10007866B2$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>Nori, Aditya Vithal</creatorcontrib><creatorcontrib>Bastani, Osbert</creatorcontrib><creatorcontrib>Criminisi, Antonio</creatorcontrib><creatorcontrib>Lampropoulos, Leonidas</creatorcontrib><creatorcontrib>Vytiniotis, Dimitrios</creatorcontrib><title>Neural network image classifier</title><description>A training engine is described which has a memory arranged to access a neural network image classifier, the neural network image classifier having been trained using a plurality of training images from an input space, the training images being labeled for a plurality of classes. The training engine has an adversarial example generator which computes a plurality of adversarial images by, for each adversarial image, searching a region in the input space around one of the training images, the region being one in which the neural network is linear, to find an image which is incorrectly classified into the plurality of classes by the neural network. The training engine has a processor which further trains the neural network image classifier using at least the adversarial images.</description><subject>CALCULATING</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2018</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZJD3Sy0tSsxRyEstKc8vylbIzE1MT1VIzkksLs5My0wt4mFgTUvMKU7lhdLcDIpuriHOHrqpBfnxqcUFicmpQK3xocGGBgYG5hZmZk5GxsSoAQA28CUr</recordid><startdate>20180626</startdate><enddate>20180626</enddate><creator>Nori, Aditya Vithal</creator><creator>Bastani, Osbert</creator><creator>Criminisi, Antonio</creator><creator>Lampropoulos, Leonidas</creator><creator>Vytiniotis, Dimitrios</creator><scope>EVB</scope></search><sort><creationdate>20180626</creationdate><title>Neural network image classifier</title><author>Nori, Aditya Vithal ; Bastani, Osbert ; Criminisi, Antonio ; Lampropoulos, Leonidas ; Vytiniotis, Dimitrios</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_US10007866B23</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng</language><creationdate>2018</creationdate><topic>CALCULATING</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>Nori, Aditya Vithal</creatorcontrib><creatorcontrib>Bastani, Osbert</creatorcontrib><creatorcontrib>Criminisi, Antonio</creatorcontrib><creatorcontrib>Lampropoulos, Leonidas</creatorcontrib><creatorcontrib>Vytiniotis, Dimitrios</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Nori, Aditya Vithal</au><au>Bastani, Osbert</au><au>Criminisi, Antonio</au><au>Lampropoulos, Leonidas</au><au>Vytiniotis, Dimitrios</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Neural network image classifier</title><date>2018-06-26</date><risdate>2018</risdate><abstract>A training engine is described which has a memory arranged to access a neural network image classifier, the neural network image classifier having been trained using a plurality of training images from an input space, the training images being labeled for a plurality of classes. The training engine has an adversarial example generator which computes a plurality of adversarial images by, for each adversarial image, searching a region in the input space around one of the training images, the region being one in which the neural network is linear, to find an image which is incorrectly classified into the plurality of classes by the neural network. The training engine has a processor which further trains the neural network image classifier using at least the adversarial images.</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_epo_espacenet_US10007866B2 |
| source | esp@cenet |
| subjects | CALCULATING COMPUTING COUNTING PHYSICS |
| title | Neural network image classifier |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-04T16%3A31%3A10IST&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=Nori,%20Aditya%20Vithal&rft.date=2018-06-26&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EUS10007866B2%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 |