{"id":"W4401211718","doi":"10.1109/isca59077.2024.00081","title":"Heterogeneous Acceleration Pipeline for Recommendation System Training","year":2024,"lang":"en","type":"article","venue":"","topic":"Education and Learning Interventions","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Acceleration; Computer science; Pipeline (software); Training (meteorology); Recommender system; Machine learning; Operating system","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003303798,0.001248559,0.0006697576,0.0004044803,0.0004902623,0.001012859,0.002028527,0.0005550015,0.02019015],"category_scores_gemma":[0.001415371,0.0005352992,0.0006584909,0.0009044605,0.0001927758,0.001393095,0.000712035,0.001600836,0.01002782],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001070193,"about_ca_system_score_gemma":0.002163027,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0204314,"about_ca_topic_score_gemma":0.03135714,"domain_scores_codex":[0.9995852,0.00003398705,0.00002301351,0.0001428291,0.0001194965,0.00009550714],"domain_scores_gemma":[0.999575,0.00007909512,0.00001672885,0.000137021,0.0001407932,0.00005142221],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001372206,0.0004974335,0.006190976,0.0003729981,0.0002308362,0.0002798541,0.0002236089,0.1036517,0.03544853,0.007407777,0.1163944,0.7279298],"study_design_scores_gemma":[0.0001797071,0.0002816973,0.002104305,0.0000212416,0.00005893888,0.0001058077,0.00006215873,0.9429301,0.02231269,0.004282051,0.02762333,0.00003798737],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09059962,0.002268544,0.7682384,0.0008646391,0.0007347799,0.0004727313,0.00363377,0.107463,0.02572458],"genre_scores_gemma":[0.5302151,0.0009306046,0.4330715,0.0006174954,0.0001537592,0.0004636799,0.008927302,0.002093838,0.0235267],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0204314,"threshold_uncertainty_score":0.06754279,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09708145438284371,"score_gpt":0.3521146602644055,"score_spread":0.2550332058815618,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}