Interpretation enhancement method based on attention mechanism
The invention discloses an interpretation enhancement method based on an attention mechanism, and relates to the field of reinforcement learning. According to the interpretation enhancement method based on the attention mechanism, a deep reinforcement learning interpretation enhancement module (IEMA...
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Patent |
| Sprache: | chi ; 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 | ZHOU XIANZHONG ZHU ZHAOQUAN SUN YUXIANG GAO BO |
| description | The invention discloses an interpretation enhancement method based on an attention mechanism, and relates to the field of reinforcement learning. According to the interpretation enhancement method based on the attention mechanism, a deep reinforcement learning interpretation enhancement module (IEMA) is included, and the deep reinforcement learning interpretation enhancement module (IEMA) comprises a channel attention module and a space attention module; the channel attention module specifically comprises the steps that global pooling is carried out on an input graph, and then the weight of an original input feature graph is obtained and acts on original input; and the space attention module specifically comprises the steps of pooling a weighted input image in a channel direction, and then carrying out convolution to obtain a space attention weight to act on input. According to the interpretation enhancement method based on the attention mechanism, a weighted feature map is obtained through combination of two |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN117610638A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN117610638A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN117610638A3</originalsourceid><addsrcrecordid>eNrjZLDzzCtJLSooSi1JLMnMz1NIzctIzEtOzU3NK1HITS3JyE9RSEosTk1RAMollpQAhUGqclOTgcoyi3N5GFjTEnOKU3mhNDeDoptriLOHbmpBfnxqcUFicmpeakm8s5-hobmZoYGZsYWjMTFqAKn3MS0</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Interpretation enhancement method based on attention mechanism</title><source>esp@cenet</source><creator>ZHOU XIANZHONG ; ZHU ZHAOQUAN ; SUN YUXIANG ; GAO BO</creator><creatorcontrib>ZHOU XIANZHONG ; ZHU ZHAOQUAN ; SUN YUXIANG ; GAO BO</creatorcontrib><description>The invention discloses an interpretation enhancement method based on an attention mechanism, and relates to the field of reinforcement learning. According to the interpretation enhancement method based on the attention mechanism, a deep reinforcement learning interpretation enhancement module (IEMA) is included, and the deep reinforcement learning interpretation enhancement module (IEMA) comprises a channel attention module and a space attention module; the channel attention module specifically comprises the steps that global pooling is carried out on an input graph, and then the weight of an original input feature graph is obtained and acts on original input; and the space attention module specifically comprises the steps of pooling a weighted input image in a channel direction, and then carrying out convolution to obtain a space attention weight to act on input. According to the interpretation enhancement method based on the attention mechanism, a weighted feature map is obtained through combination of two</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; PHYSICS</subject><creationdate>2024</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=20240227&DB=EPODOC&CC=CN&NR=117610638A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,776,881,25542,76290</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240227&DB=EPODOC&CC=CN&NR=117610638A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>ZHOU XIANZHONG</creatorcontrib><creatorcontrib>ZHU ZHAOQUAN</creatorcontrib><creatorcontrib>SUN YUXIANG</creatorcontrib><creatorcontrib>GAO BO</creatorcontrib><title>Interpretation enhancement method based on attention mechanism</title><description>The invention discloses an interpretation enhancement method based on an attention mechanism, and relates to the field of reinforcement learning. According to the interpretation enhancement method based on the attention mechanism, a deep reinforcement learning interpretation enhancement module (IEMA) is included, and the deep reinforcement learning interpretation enhancement module (IEMA) comprises a channel attention module and a space attention module; the channel attention module specifically comprises the steps that global pooling is carried out on an input graph, and then the weight of an original input feature graph is obtained and acts on original input; and the space attention module specifically comprises the steps of pooling a weighted input image in a channel direction, and then carrying out convolution to obtain a space attention weight to act on input. According to the interpretation enhancement method based on the attention mechanism, a weighted feature map is obtained through combination of two</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZLDzzCtJLSooSi1JLMnMz1NIzctIzEtOzU3NK1HITS3JyE9RSEosTk1RAMollpQAhUGqclOTgcoyi3N5GFjTEnOKU3mhNDeDoptriLOHbmpBfnxqcUFicmpeakm8s5-hobmZoYGZsYWjMTFqAKn3MS0</recordid><startdate>20240227</startdate><enddate>20240227</enddate><creator>ZHOU XIANZHONG</creator><creator>ZHU ZHAOQUAN</creator><creator>SUN YUXIANG</creator><creator>GAO BO</creator><scope>EVB</scope></search><sort><creationdate>20240227</creationdate><title>Interpretation enhancement method based on attention mechanism</title><author>ZHOU XIANZHONG ; ZHU ZHAOQUAN ; SUN YUXIANG ; GAO BO</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN117610638A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2024</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>ZHOU XIANZHONG</creatorcontrib><creatorcontrib>ZHU ZHAOQUAN</creatorcontrib><creatorcontrib>SUN YUXIANG</creatorcontrib><creatorcontrib>GAO BO</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>ZHOU XIANZHONG</au><au>ZHU ZHAOQUAN</au><au>SUN YUXIANG</au><au>GAO BO</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Interpretation enhancement method based on attention mechanism</title><date>2024-02-27</date><risdate>2024</risdate><abstract>The invention discloses an interpretation enhancement method based on an attention mechanism, and relates to the field of reinforcement learning. According to the interpretation enhancement method based on the attention mechanism, a deep reinforcement learning interpretation enhancement module (IEMA) is included, and the deep reinforcement learning interpretation enhancement module (IEMA) comprises a channel attention module and a space attention module; the channel attention module specifically comprises the steps that global pooling is carried out on an input graph, and then the weight of an original input feature graph is obtained and acts on original input; and the space attention module specifically comprises the steps of pooling a weighted input image in a channel direction, and then carrying out convolution to obtain a space attention weight to act on input. According to the interpretation enhancement method based on the attention mechanism, a weighted feature map is obtained through combination of two</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN117610638A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING PHYSICS |
| title | Interpretation enhancement method based on attention mechanism |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-01T16%3A43%3A09IST&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=ZHOU%20XIANZHONG&rft.date=2024-02-27&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN117610638A%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 |