{"id":"https://openalex.org/W2891584070","doi":"https://doi.org/10.1109/icassp.2018.8461511","title":"Transferring Information Between Neural Networks","display_name":"Transferring Information Between Neural Networks","publication_year":2018,"publication_date":"2018-04-01","ids":{"openalex":"https://openalex.org/W2891584070","doi":"https://doi.org/10.1109/icassp.2018.8461511","mag":"2891584070"},"language":"en","primary_location":{"id":"doi:10.1109/icassp.2018.8461511","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2018.8461511","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5076439728","display_name":"Christopher Ehmann","orcid":null},"institutions":[{"id":"https://openalex.org/I2800274787","display_name":"Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute","ror":"https://ror.org/02tbr6331","country_code":"DE","type":"facility","lineage":["https://openalex.org/I2800274787","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Christopher Ehmann","raw_affiliation_strings":["Fraunhofer Heinrich Hertz Institute, Machine Learning Group, Berlin, Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer Heinrich Hertz Institute, Machine Learning Group, Berlin, Germany","institution_ids":["https://openalex.org/I2800274787"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5026451495","display_name":"Wojciech Samek","orcid":"https://orcid.org/0000-0002-6283-3265"},"institutions":[{"id":"https://openalex.org/I2800274787","display_name":"Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute","ror":"https://ror.org/02tbr6331","country_code":"DE","type":"facility","lineage":["https://openalex.org/I2800274787","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Wojciech Samek","raw_affiliation_strings":["Fraunhofer Heinrich Hertz Institute, Machine Learning Group, Berlin, Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer Heinrich Hertz Institute, Machine Learning Group, Berlin, Germany","institution_ids":["https://openalex.org/I2800274787"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5076439728"],"corresponding_institution_ids":["https://openalex.org/I2800274787"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.09750708,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":96},"biblio":{"volume":"17","issue":null,"first_page":"2361","last_page":"2365"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9995999932289124,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9995999932289124,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9993000030517578,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9983999729156494,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/softmax-function","display_name":"Softmax function","score":0.8523320555686951},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.8247570991516113},{"id":"https://openalex.org/keywords/jacobian-matrix-and-determinant","display_name":"Jacobian matrix and determinant","score":0.7445736527442932},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.7119473218917847},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6914777755737305},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.5692648887634277},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.44894787669181824},{"id":"https://openalex.org/keywords/mean-squared-error","display_name":"Mean squared error","score":0.4320053458213806},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3841431736946106},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.21481925249099731},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.12057706713676453},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.11307653784751892}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.8523320555686951},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.8247570991516113},{"id":"https://openalex.org/C200331156","wikidata":"https://www.wikidata.org/wiki/Q506041","display_name":"Jacobian matrix and determinant","level":2,"score":0.7445736527442932},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.7119473218917847},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6914777755737305},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.5692648887634277},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44894787669181824},{"id":"https://openalex.org/C139945424","wikidata":"https://www.wikidata.org/wiki/Q1940696","display_name":"Mean squared error","level":2,"score":0.4320053458213806},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3841431736946106},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.21481925249099731},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.12057706713676453},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.11307653784751892}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/icassp.2018.8461511","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2018.8461511","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},{"id":"pmh:oai:fraunhofer.de:N-520115","is_oa":false,"landing_page_url":"http://publica.fraunhofer.de/documents/N-520115.html","pdf_url":null,"source":{"id":"https://openalex.org/S4306400801","display_name":"Publikationsdatenbank der Fraunhofer-Gesellschaft (Fraunhofer-Gesellschaft)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4923324","host_organization_name":"Fraunhofer-Gesellschaft","host_organization_lineage":["https://openalex.org/I4923324"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Fraunhofer HHI","raw_type":"Conference Paper"},{"id":"pmh:oai:publica.fraunhofer.de:publica/403483","is_oa":false,"landing_page_url":"https://publica.fraunhofer.de/handle/publica/403483","pdf_url":null,"source":{"id":"https://openalex.org/S4306400318","display_name":"Fraunhofer-Publica (Fraunhofer-Gesellschaft)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4923324","host_organization_name":"Fraunhofer-Gesellschaft","host_organization_lineage":["https://openalex.org/I4923324"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"conference paper"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.7400000095367432,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W1787224781","https://openalex.org/W1821462560","https://openalex.org/W2112594540","https://openalex.org/W2119144962","https://openalex.org/W2157331557","https://openalex.org/W2162390675","https://openalex.org/W2194775991","https://openalex.org/W2195388612","https://openalex.org/W2240067561","https://openalex.org/W2257979135","https://openalex.org/W2336509392","https://openalex.org/W2561238782","https://openalex.org/W2563786098","https://openalex.org/W2606722458","https://openalex.org/W2657631929","https://openalex.org/W2953384591","https://openalex.org/W2962851944","https://openalex.org/W2963340555","https://openalex.org/W2963798744","https://openalex.org/W2964299589","https://openalex.org/W3005000804","https://openalex.org/W3100498948","https://openalex.org/W6683258052","https://openalex.org/W6683722107","https://openalex.org/W6713134421","https://openalex.org/W6727323963","https://openalex.org/W6734692528"],"related_works":["https://openalex.org/W3107204728","https://openalex.org/W4287591324","https://openalex.org/W3108503355","https://openalex.org/W4226420367","https://openalex.org/W2962876041","https://openalex.org/W3090555870","https://openalex.org/W3022820045","https://openalex.org/W2801655600","https://openalex.org/W3005627584","https://openalex.org/W4303493643"],"abstract_inverted_index":{"This":[0,158],"paper":[1],"investigates":[2],"techniques":[3],"to":[4,20,42,54,63,96,117,125],"transfer":[5,162],"information":[6,21,161],"between":[7,76,134],"deep":[8,36],"neural":[9],"networks.":[10],"We":[11,67],"demonstrate":[12],"that":[13,49,160],"a":[14,24,44,60,102,164],"student":[15,53,137],"network,":[16],"which":[17],"has":[18,163],"access":[19],"computed":[22],"by":[23,58],"teacher":[25,57,135],"network":[26],"on":[27],"the":[28,52,56,64,77,81,84,92,114,118,126,130,146,154],"training":[29,121,156],"data,":[30],"learns":[31],"faster,":[32],"can":[33],"be":[34],"less":[35,39],"and":[37,100,136],"requires":[38],"labeled":[40],"examples":[41],"achieve":[43],"given":[45],"performance":[46],"level.":[47],"For":[48],"we":[50,141],"force":[51],"mimic":[55],"adding":[59],"penalty":[61,70],"term":[62],"student's":[65],"objective.":[66],"evaluate":[68],"different":[69],"terms:":[71],"(1)":[72],"mean":[73],"squared":[74],"error":[75],"cost":[78,93,127],"gradients,":[79],"(2)":[80],"Jacobian":[82,111,147],"of":[83],"pre-softmax":[85],"layer,":[86],"(3)":[87],"its":[88],"row-summed":[89],"version,":[90],"(4)":[91],"gradient":[94,107],"differences":[95],"standard":[97],"double":[98,104],"backpropagation":[99,105],"(5)":[101],"targeted":[103],"via":[106],"derived":[108],"masks.":[109],"The":[110],"method":[112],"improves":[113],"accuracy":[115,133],"proportional":[116],"difference":[119,131],"in":[120,123,132],"examples,":[122],"contrast":[124],"gradient.":[128],"If":[129],"is":[138],"large":[139],"enough,":[140],"find":[142],"an":[143],"improvement":[144],"from":[145],"information,":[148],"even":[149],"if":[150],"both":[151],"had":[152],"seen":[153],"same":[155],"data.":[157],"indicates":[159],"regularization":[165],"effect.":[166]},"counts_by_year":[{"year":2023,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}