{"id":"W3200211247","doi":"10.14778/3485450.3485462","title":"Accelerating recommendation system training by leveraging popular choices","year":2021,"lang":"en","type":"article","venue":"Proceedings of the VLDB Endowment","topic":"Recommender Systems and Techniques","field":"Computer Science","cited_by":50,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Embedding; Recommender system; Categorical variable; Popularity; Parallel computing; Feature (linguistics); Representation (politics); Machine learning; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006726156,0.000157435,0.0002425817,0.00005552347,0.000247206,0.0003405347,0.0007558118,0.00004907969,0.000007969791],"category_scores_gemma":[0.00004223103,0.0001204821,0.0001042016,0.0003474764,0.00001264734,0.0006351399,0.0004596283,0.0001446551,0.000001446392],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001426342,"about_ca_system_score_gemma":0.00003347871,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009116113,"about_ca_topic_score_gemma":0.000002209611,"domain_scores_codex":[0.9985837,0.00002690906,0.0004467211,0.0003824806,0.0002984006,0.0002618374],"domain_scores_gemma":[0.9991394,0.00002789698,0.0004062193,0.0001832992,0.0001904704,0.00005266416],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000007038718,0.0002415237,0.01201409,0.001253439,0.0002666787,0.00000395806,0.01286651,0.00001150088,0.4108203,0.2821753,0.03233436,0.2480053],"study_design_scores_gemma":[0.0005604162,0.00006551785,0.0003677594,0.0008660727,0.0000260955,0.0001381244,0.004564815,0.01055325,0.9260672,0.001820292,0.05456804,0.0004023781],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4552433,0.002822582,0.2709108,0.03699035,0.007021845,0.003868476,0.00004657478,0.003192077,0.219904],"genre_scores_gemma":[0.9739036,0.00001782071,0.02546209,0.0002011604,0.00007656005,0.00006692087,0.000003305381,0.00001283631,0.0002556922],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5186602,"threshold_uncertainty_score":0.4913116,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04706617408481114,"score_gpt":0.2487423492667841,"score_spread":0.2016761751819729,"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."}}