{"id":"https://openalex.org/W4413146471","doi":"https://doi.org/10.1109/cvpr52734.2025.01933","title":"Rate-In: Information-Driven Adaptive Dropout Rates for Improved Inference-Time Uncertainty Estimation","display_name":"Rate-In: Information-Driven Adaptive Dropout Rates for Improved Inference-Time Uncertainty Estimation","publication_year":2025,"publication_date":"2025-06-10","ids":{"openalex":"https://openalex.org/W4413146471","doi":"https://doi.org/10.1109/cvpr52734.2025.01933"},"language":"en","primary_location":{"id":"doi:10.1109/cvpr52734.2025.01933","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52734.2025.01933","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","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/A5064608267","display_name":"Tal Zeevi","orcid":"https://orcid.org/0000-0001-8917-7061"},"institutions":[{"id":"https://openalex.org/I32971472","display_name":"Yale University","ror":"https://ror.org/03v76x132","country_code":"US","type":"education","lineage":["https://openalex.org/I32971472"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Tal Zeevi","raw_affiliation_strings":["Yale University"],"affiliations":[{"raw_affiliation_string":"Yale University","institution_ids":["https://openalex.org/I32971472"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036015811","display_name":"Ravid Shwartz-Ziv","orcid":null},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ravid Shwartz-Ziv","raw_affiliation_strings":["New York University"],"affiliations":[{"raw_affiliation_string":"New York University","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001226970","display_name":"Yann LeCun","orcid":null},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yann LeCun","raw_affiliation_strings":["New York University"],"affiliations":[{"raw_affiliation_string":"New York University","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078494006","display_name":"Lawrence H. Staib","orcid":"https://orcid.org/0000-0002-9516-5136"},"institutions":[{"id":"https://openalex.org/I32971472","display_name":"Yale University","ror":"https://ror.org/03v76x132","country_code":"US","type":"education","lineage":["https://openalex.org/I32971472"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lawrence H. Staib","raw_affiliation_strings":["Yale University"],"affiliations":[{"raw_affiliation_string":"Yale University","institution_ids":["https://openalex.org/I32971472"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5083553196","display_name":"John A. Onofrey","orcid":"https://orcid.org/0000-0002-9432-0448"},"institutions":[{"id":"https://openalex.org/I32971472","display_name":"Yale University","ror":"https://ror.org/03v76x132","country_code":"US","type":"education","lineage":["https://openalex.org/I32971472"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"John A. Onofrey","raw_affiliation_strings":["Yale University"],"affiliations":[{"raw_affiliation_string":"Yale University","institution_ids":["https://openalex.org/I32971472"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5064608267"],"corresponding_institution_ids":["https://openalex.org/I32971472"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.28118037,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"20757","last_page":"20766"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11195","display_name":"Simulation Techniques and Applications","score":0.9314000010490417,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T11195","display_name":"Simulation Techniques and Applications","score":0.9314000010490417,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/dropout","display_name":"Dropout (neural networks)","score":0.8307013511657715},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6552340984344482},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6517614722251892},{"id":"https://openalex.org/keywords/estimation","display_name":"Estimation","score":0.5818284749984741},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.35559916496276855},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3244069814682007},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08746486902236938}],"concepts":[{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.8307013511657715},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6552340984344482},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6517614722251892},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.5818284749984741},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35559916496276855},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3244069814682007},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08746486902236938},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cvpr52734.2025.01933","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52734.2025.01933","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W4387369504","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W3107602296","https://openalex.org/W4364306694","https://openalex.org/W4312192474","https://openalex.org/W4283697347"],"abstract_inverted_index":{"Accurate":[0],"uncertainty":[1,27,51,91,195,222],"estimation":[2,223],"is":[3,19],"crucial":[4],"for":[5,24,218],"deploying":[6],"neural":[7],"networks":[8],"in":[9,79,93,119,156,224],"risk-sensitive":[10],"applications":[11],"such":[12],"as":[13,53,127],"medical":[14,170,185],"diagnosis.":[15],"Monte":[16,94],"Carlo":[17,95],"Dropout":[18],"a":[20,210],"widely":[21],"used":[22],"technique":[23],"approximating":[25],"predictive":[26,206,221],"by":[28,111,117],"performing":[29],"stochastic":[30],"forward":[31],"passes":[32],"with":[33],"dropout":[34,40,71,107,118,126,137,162,202,217],"during":[35,73,109],"inference.":[36],"However,":[37],"using":[38,75],"static":[39],"rates":[41,72,108,138,163,203],"across":[42,168],"all":[43],"layers":[44],"and":[45,65,131,141,173,183,193],"inputs":[46,64],"can":[47],"lead":[48],"to":[49,56,58,86,164,198,215],"suboptimal":[50],"estimates,":[52],"it":[54],"fails":[55],"adapt":[57],"the":[59,113,152],"varying":[60],"characteristics":[61],"of":[62],"individual":[63],"network":[66],"layers.":[67],"Existing":[68],"approaches":[69],"optimize":[70],"training":[74],"labeled":[76],"data,":[77],"resulting":[78],"fixed":[80,199],"inference-time":[81,213],"parameters":[82],"that":[83,104,189],"cannot":[84],"adjust":[85],"new":[87],"data":[88,182],"distributions,":[89],"compromising":[90,205],"estimates":[92,196],"simulations.In":[96],"this":[97],"paper,":[98],"we":[99,159],"propose":[100],"Rate-In,":[101],"an":[102],"algorithm":[103],"dynamically":[105],"adjusts":[106],"inference":[110],"quantifying":[112,151],"information":[114,154],"loss":[115,155],"induced":[116],"each":[120],"layer\u2019s":[121],"feature":[122,157],"maps.":[123],"By":[124,150],"treating":[125],"controlled":[128],"noise":[129],"injection":[130],"leveraging":[132],"information-theoretic":[133],"principles,":[134],"Rate-In":[135,190,208],"adapts":[136],"per":[139,142],"layer":[140],"input":[143],"instance":[144],"without":[145,204],"requiring":[146],"ground":[147],"truth":[148],"labels.":[149],"functional":[153],"maps,":[158],"adaptively":[160],"tune":[161],"maintain":[165],"perceptual":[166],"quality":[167],"diverse":[169],"imaging":[171,186],"tasks":[172,187],"architectural":[174],"configurations.":[175],"Our":[176],"extensive":[177],"empirical":[178],"study":[179],"on":[180],"synthetic":[181],"real-world":[184],"demonstrates":[188],"improves":[191],"calibration":[192],"sharpens":[194],"compared":[197],"or":[200],"heuristic":[201],"performance.":[207],"offers":[209],"practical,":[211],"unsupervised,":[212],"approach":[214],"optimizing":[216],"more":[219],"reliable":[220],"critical":[225],"applications.":[226]},"counts_by_year":[],"updated_date":"2025-12-28T23:10:05.387466","created_date":"2025-10-10T00:00:00"}