{"id":"https://openalex.org/W7151509701","doi":"https://doi.org/10.48550/arxiv.2604.04736","title":"Sampling Parallelism for Fast and Efficient Bayesian Learning","display_name":"Sampling Parallelism for Fast and Efficient Bayesian Learning","publication_year":2026,"publication_date":"2026-04-06","ids":{"openalex":"https://openalex.org/W7151509701","doi":"https://doi.org/10.48550/arxiv.2604.04736"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.04736","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.04736","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.04736","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5123382242","display_name":"Asena Karolin \u00d6zdemir","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"\u00d6zdemir, Asena Karolin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133085271","display_name":"Lars H. Heyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Heyen, Lars H.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010580917","display_name":"Arvid Weyrauch","orcid":"https://orcid.org/0000-0002-2684-0927"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Weyrauch, Arvid","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030034618","display_name":"Achim Streit","orcid":"https://orcid.org/0000-0002-5065-469X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Streit, Achim","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067996051","display_name":"Markus G\u00f6tz","orcid":"https://orcid.org/0000-0002-2233-1041"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"G\u00f6tz, Markus","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5024295845","display_name":"Charlotte Debus","orcid":"https://orcid.org/0000-0002-7156-2022"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Debus, Charlotte","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":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.2924000024795532,"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"}},"topics":[{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.2924000024795532,"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/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.17080000042915344,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.11819999665021896,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.7337999939918518},{"id":"https://openalex.org/keywords/hyperparameter","display_name":"Hyperparameter","score":0.6414999961853027},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.5824999809265137},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.5662999749183655},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.5009999871253967},{"id":"https://openalex.org/keywords/parallelism","display_name":"Parallelism (grammar)","score":0.48080000281333923},{"id":"https://openalex.org/keywords/bayesian-network","display_name":"Bayesian network","score":0.4424999952316284},{"id":"https://openalex.org/keywords/data-parallelism","display_name":"Data parallelism","score":0.41190001368522644},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.38909998536109924}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7958999872207642},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.7337999939918518},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.6414999961853027},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.5824999809265137},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.5662999749183655},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5546000003814697},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5439000129699707},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.5009999871253967},{"id":"https://openalex.org/C2781172179","wikidata":"https://www.wikidata.org/wiki/Q853109","display_name":"Parallelism (grammar)","level":2,"score":0.48080000281333923},{"id":"https://openalex.org/C33724603","wikidata":"https://www.wikidata.org/wiki/Q812540","display_name":"Bayesian network","level":2,"score":0.4424999952316284},{"id":"https://openalex.org/C61483411","wikidata":"https://www.wikidata.org/wiki/Q3124522","display_name":"Data parallelism","level":3,"score":0.41190001368522644},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.38909998536109924},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.34380000829696655},{"id":"https://openalex.org/C73602740","wikidata":"https://www.wikidata.org/wiki/Q7795822","display_name":"Thompson sampling","level":3,"score":0.33090001344680786},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.32710000872612},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.320499986410141},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.31929999589920044},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3165000081062317},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.3077999949455261},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.30550000071525574},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.3028999865055084},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.28450000286102295},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.26739999651908875},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26589998602867126},{"id":"https://openalex.org/C140763907","wikidata":"https://www.wikidata.org/wiki/Q2714055","display_name":"Instruction-level parallelism","level":3,"score":0.2551000118255615},{"id":"https://openalex.org/C101112237","wikidata":"https://www.wikidata.org/wiki/Q4874481","display_name":"Bayesian statistics","level":4,"score":0.2535000145435333}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.04736","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.04736","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.04736","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.04736","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Machine":[0],"learning":[1,48],"models,":[2],"and":[3,20,61,69,78,123,138,147,177,235],"deep":[4],"neural":[5,53],"networks":[6,54],"in":[7,12],"particular,":[8],"are":[9,56],"increasingly":[10],"deployed":[11],"risk-sensitive":[13],"domains":[14],"such":[15,50],"as":[16,51,155],"healthcare,":[17],"environmental":[18],"forecasting,":[19],"finance,":[21],"where":[22],"reliable":[23],"quantification":[24,33],"of":[25,80,104,240],"predictive":[26],"uncertainty":[27,32],"is":[28,164,189],"essential.":[29],"However,":[30],"many":[31],"(UQ)":[34],"methods":[35],"remain":[36],"difficult":[37],"to":[38,41,166,192,224],"apply":[39],"due":[40],"their":[42],"substantial":[43],"computational":[44,194],"cost.":[45],"Sampling-based":[46],"Bayesian":[47,52,81,106],"approaches,":[49],"(BNNs),":[55],"particularly":[57],"expensive":[58],"since":[59],"drawing":[60],"evaluating":[62],"multiple":[63,116],"parameter":[64],"samples":[65,109],"rapidly":[66],"exhausts":[67],"memory":[68,121],"compute":[70],"resources.":[71],"These":[72],"constraints":[73],"have":[74],"limited":[75],"the":[76,101,108,136,186,193,225,238],"accessibility":[77],"exploration":[79],"techniques":[82],"thus":[83,236],"far.":[84],"To":[85],"address":[86],"these":[87],"challenges,":[88],"we":[89],"introduce":[90],"sampling":[91,162,213],"parallelism,":[92],"a":[93,143,156,171,216],"simple":[94],"yet":[95],"powerful":[96],"parallelization":[97],"strategy":[98],"that":[99,161,174,197],"targets":[100],"primary":[102],"bottleneck":[103],"sampling-based":[105],"learning:":[107],"themselves.":[110],"By":[111],"distributing":[112],"sample":[113,176,187,198],"evaluations":[114,199],"across":[115],"GPUs,":[117],"our":[118],"method":[119],"reduces":[120,237],"pressure":[122],"training":[124],"time":[125],"without":[126],"requiring":[127],"architectural":[128],"changes":[129],"or":[130],"extensive":[131],"hyperparameter":[132],"tuning.":[133],"We":[134,158],"detail":[135],"methodology":[137],"evaluate":[139],"its":[140],"performance":[141],"on":[142,228],"few":[144],"example":[145],"tasks":[146],"architectures,":[148],"comparing":[149],"against":[150],"distributed":[151],"data":[152,178],"parallelism":[153,163,214],"(DDP)":[154],"baseline.":[157],"further":[159],"demonstrate":[160],"complementary":[165],"existing":[167],"strategies":[168],"by":[169,219],"implementing":[170],"hybrid":[172],"approach":[173],"combines":[175],"parallelism.":[179],"Our":[180],"experiments":[181],"show":[182],"near-perfect":[183],"scaling":[184,209],"when":[185],"number":[188,239],"scaled":[190],"proportionally":[191],"resources,":[195],"confirming":[196],"parallelize":[200],"cleanly.":[201],"Although":[202],"DDP":[203],"achieves":[204],"better":[205],"raw":[206],"speedups":[207],"under":[208],"with":[210],"constant":[211],"workload,":[212],"has":[215],"notable":[217],"advantage:":[218],"applying":[220],"independent":[221],"stochastic":[222],"augmentations":[223],"same":[226],"batch":[227],"each":[229],"GPU,":[230],"it":[231],"increases":[232],"augmentation":[233],"diversity":[234],"epochs":[241],"required":[242],"for":[243],"convergence.":[244]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-04-08T00:00:00"}