{"id":"https://openalex.org/W6959523841","doi":"https://doi.org/10.13016/ceqz-gfrx","title":"Effective Training and Efficient Inference of Deep Neural Networks for Visual Understanding","display_name":"Effective Training and Efficient Inference of Deep Neural Networks for Visual Understanding","publication_year":2022,"publication_date":"2022-01-01","ids":{"openalex":"https://openalex.org/W6959523841","doi":"https://doi.org/10.13016/ceqz-gfrx"},"language":"en","primary_location":{"id":"doi:10.13016/ceqz-gfrx","is_oa":true,"landing_page_url":"https://doi.org/10.13016/ceqz-gfrx","pdf_url":null,"source":{"id":"https://openalex.org/S4306402644","display_name":"Digital Repository at the University of Maryland (University of Maryland College Park)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I66946132","host_organization_name":"University of Maryland, College Park","host_organization_lineage":["https://openalex.org/I66946132"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Collection"},"type":"other","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.13016/ceqz-gfrx","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Li, Hengduo","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Li, Hengduo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T11468","display_name":"Genetic Mapping and Diversity in Plants and Animals","score":0.22769999504089355,"subfield":{"id":"https://openalex.org/subfields/1311","display_name":"Genetics"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T11468","display_name":"Genetic Mapping and Diversity in Plants and Animals","score":0.22769999504089355,"subfield":{"id":"https://openalex.org/subfields/1311","display_name":"Genetics"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T12045","display_name":"Rice Cultivation and Yield Improvement","score":0.14000000059604645,"subfield":{"id":"https://openalex.org/subfields/1110","display_name":"Plant Science"},"field":{"id":"https://openalex.org/fields/11","display_name":"Agricultural and Biological Sciences"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10012","display_name":"Genetic diversity and population structure","score":0.10830000042915344,"subfield":{"id":"https://openalex.org/subfields/1311","display_name":"Genetics"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7853999733924866},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.5214999914169312},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4871000051498413},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.4715999960899353},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.46779999136924744},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.42879998683929443},{"id":"https://openalex.org/keywords/annotation","display_name":"Annotation","score":0.415800005197525},{"id":"https://openalex.org/keywords/cognitive-neuroscience-of-visual-object-recognition","display_name":"Cognitive neuroscience of visual object recognition","score":0.413100004196167},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.4009999930858612}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8134999871253967},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7853999733924866},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7010999917984009},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6502000093460083},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5214999914169312},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4871000051498413},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.4715999960899353},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.46779999136924744},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.42879998683929443},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.415800005197525},{"id":"https://openalex.org/C64876066","wikidata":"https://www.wikidata.org/wiki/Q5141226","display_name":"Cognitive neuroscience of visual object recognition","level":3,"score":0.413100004196167},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.4009999930858612},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.40049999952316284},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.39089998602867126},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.3628000020980835},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.35429999232292175},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.3422999978065491},{"id":"https://openalex.org/C186108316","wikidata":"https://www.wikidata.org/wiki/Q352530","display_name":"Adaptive neuro fuzzy inference system","level":4,"score":0.33820000290870667},{"id":"https://openalex.org/C77660652","wikidata":"https://www.wikidata.org/wiki/Q150971","display_name":"Computer graphics","level":2,"score":0.3172000050544739},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.2964000105857849},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.27720001339912415},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.27000001072883606},{"id":"https://openalex.org/C66905080","wikidata":"https://www.wikidata.org/wiki/Q17005494","display_name":"Binary classification","level":3,"score":0.2678999900817871},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.26510000228881836},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.258899986743927},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.2581000030040741},{"id":"https://openalex.org/C205711294","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Rendering (computer graphics)","level":2,"score":0.2572000026702881},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.25279998779296875},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.25270000100135803}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.13016/ceqz-gfrx","is_oa":true,"landing_page_url":"https://doi.org/10.13016/ceqz-gfrx","pdf_url":null,"source":{"id":"https://openalex.org/S4306402644","display_name":"Digital Repository at the University of Maryland (University of Maryland College Park)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I66946132","host_organization_name":"University of Maryland, College Park","host_organization_lineage":["https://openalex.org/I66946132"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Collection"}],"best_oa_location":{"id":"doi:10.13016/ceqz-gfrx","is_oa":true,"landing_page_url":"https://doi.org/10.13016/ceqz-gfrx","pdf_url":null,"source":{"id":"https://openalex.org/S4306402644","display_name":"Digital Repository at the University of Maryland (University of Maryland College Park)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I66946132","host_organization_name":"University of Maryland, College Park","host_organization_lineage":["https://openalex.org/I66946132"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Collection"},"sustainable_development_goals":[{"display_name":"Decent work and economic growth","score":0.5105472207069397,"id":"https://metadata.un.org/sdg/8"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Since":[0],"the":[1,12,45,70,82,93,105,115,121,163,174,186,201,207,230,257,281,288,297,303,306,386,417],"phenomenal":[2],"success":[3],"of":[4,28,51,95,117,124,177,214,276,283,305,374,388],"deep":[5],"neural":[6],"networks":[7,21,180],"(DNNs)":[8],"on":[9,308,313,377,406],"image":[10,314,395],"classification,":[11],"research":[13],"community":[14],"have":[15],"been":[16],"developing":[17,143],"wider":[18],"and":[19,63,68,113,148,157,181,189,209,220,233,244,262,287,316,322,349,352,369,397,411,431],"deeper":[20],"with":[22,76,229,238,256],"complex":[23],"components":[24],"for":[25,153,185,325,343,381,394,428],"a":[26,48,226,378,437],"variety":[27],"visual":[29],"understanding":[30],"tasks.":[31],"While":[32],"such":[33,222],"\u201cheavy\u201d":[34],"models":[35,159,376],"achieve":[36],"excellent":[37],"performance,":[38],"they":[39],"pose":[40],"two":[41,168,273],"main":[42],"challenges:":[43],"(1)":[44],"training":[46,96,154,215,236],"requires":[47],"significant":[49],"amount":[50],"computational":[52,122,347,361,439],"resource":[53],"as":[54,56],"well":[55],"large-scale":[57],"labeled":[58],"datasets":[59],"acquired":[60],"from":[61,206],"time-consuming":[62],"labor-intensive":[64],"human":[65],"annotation":[66,109],"process;":[67],"(2)":[69],"inference":[71,118,175,331,341],"can":[72],"be":[73],"slow":[74],"even":[75],"expensive":[77],"graphics":[78],"cards":[79],"due":[80],"to":[81,251,358,414,424],"high":[83,128],"model":[84],"complexity.":[85],"To":[86,329],"address":[87],"these":[88],"challenges,":[89],"we":[90,133,166,193,199,300,333],"explore":[91,334],"improving":[92,114],"effectiveness":[94],"DNNs":[97,125],"so":[98],"that":[99,119,172,269,338],"better":[100,144],"performance":[101,255],"is":[102],"achieved":[103],"under":[104],"same":[106],"computation":[107,170,336,421,427],"and/or":[108],"cost":[110,123],"during":[111],"training,":[112],"efficiency":[116,176,400],"reduces":[120],"while":[126],"maintaining":[127],"accuracy.":[129],"In":[130,162,296,354],"this":[131],"dissertation,":[132],"first":[134,194],"propose":[135,263],"several":[136],"approaches":[137],"including":[138],"devising":[139],"noise-aware":[140],"supervisory":[141],"signals,":[142],"semi-supervised":[145,265,277],"learning":[146,266],"methods":[147,337,422],"analyzing":[149],"different":[150],"pre-training":[151,312],"techniques":[152,268],"object":[155,218,253,278,317],"recognition":[156],"detection":[158,254],"more":[160,320],"effectively.":[161],"second":[164],"part,":[165],"present":[167,350],"adaptive":[169,335,420],"frameworks":[171],"improve":[173,330],"3D":[178,371,375],"convolutional":[179,372],"attention-based":[182],"vision":[183],"Transformers":[184,393],"tasks":[187,310],"ofimage":[188],"video":[190,382],"classification.":[191],"Specifically,":[192],"introduce":[195],"NoisyAnchor,":[196],"in":[197,217,391,436],"which":[198,407],"identify":[200],"intrinsic":[202],"label":[203],"noise":[204,223],"generated":[205],"harsh":[208],"binary":[210],"IoU-based":[211],"(Intersection-over-Union)foreground/background":[212],"split":[213],"samples":[216,237],"detection,":[219,279,318],"mitigate":[221],"through":[224,401],"deriving":[225,402],"cleanliness":[227],"score":[228],"detectors\u2019":[231],"output":[232],"down-weight":[234],"noisy":[235],"further":[239],"derived":[240],"soft":[241],"category":[242],"labels":[243],"loss":[245],"re-weighting":[246],"coefficients.":[247],"We":[248],"then":[249],"seek":[250],"boost":[252],"readily":[258],"available":[259],"unannotated":[260],"images,":[261],"improved":[264],"(SSL)":[267],"aim":[270],"at":[271],"addressing":[272],"unique":[274],"challenges":[275],"i.e.":[280],"lack":[282],"localization":[284],"quality":[285],"estimation":[286],"amplified":[289],"class":[290],"imbalance":[291],"when":[292,311],"generating":[293],"pseudo":[294],"labels.":[295],"third":[298],"work":[299],"empirically":[301],"analyze":[302],"differences":[304],"impact":[307],"downstream":[309],"classification":[315,396],"providing":[319],"intuitions":[321],"actionable":[323],"practices":[324],"effective":[326],"task-specific":[327],"pre-training.":[328],"efficiency,":[332],"produce":[339],"input-specific":[340,403],"policies":[342,405],"an":[344],"overall":[345],"reduced":[346,438],"cost,":[348],"Ada3D":[351,356],"AdaViT.":[353],"particular,":[355],"learns":[357],"adaptively":[359],"allocate":[360,425],"resources":[362],"by":[363],"selectively":[364],"keeping":[365],"informative":[366],"input":[367],"frames":[368],"activating":[370],"layers":[373],"per-input":[379],"basis":[380],"classification;":[383],"AdaViT":[384],"exploits":[385],"redundancy":[387],"self-attention":[389,409],"mechanism":[390],"Vision":[392],"improves":[398],"their":[399],"usage":[404],"patches,":[408],"heads":[410],"transformer":[412],"blocks":[413],"use":[415],"throughout":[416],"backbone.":[418],"Such":[419],"tend":[423],"less":[426],"\u201ceasy\u201d":[429],"images":[430],"\u201cstatic\u201d":[432],"videos,":[433],"thus":[434],"resulting":[435],"cost.":[440]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}