{"id":"W1972783048","doi":"10.1145/2688500.2688521","title":"On optimizing machine learning workloads via kernel fusion","year":2015,"lang":"en","type":"article","venue":"","topic":"Parallel Computing and Optimization Techniques","field":"Computer Science","cited_by":40,"is_retracted":false,"has_abstract":true,"ca_institutions":"IBM (Canada)","funders":"","keywords":"Computer science; Kernel (algebra); Scalability; Locality; Matrix multiplication; Parallel computing; Computation; Operator (biology); Range (aeronautics); Algorithm; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008092055,0.0007489531,0.0006369162,0.0004077291,0.0004242382,0.0008186869,0.000892349,0.0004922295,0.001255994],"category_scores_gemma":[0.004105197,0.000241564,0.0003623886,0.001022816,0.0004989136,0.001721038,0.0009262419,0.0007694774,0.0006003498],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007545933,"about_ca_system_score_gemma":0.001304029,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002491873,"about_ca_topic_score_gemma":0.003052617,"domain_scores_codex":[0.9992513,0.0001972253,0.00004236579,0.00009563388,0.000285515,0.0001278204],"domain_scores_gemma":[0.9986035,0.0006719845,0.00009298832,0.0002895347,0.0002783857,0.00006356771],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002147638,0.0001659429,0.001634658,0.00006721754,0.00002857209,0.0000605002,0.00009252845,0.8232573,0.01902404,0.01214389,0.001926038,0.1413846],"study_design_scores_gemma":[0.000003576667,0.00001577938,0.00009051721,0.000001375907,0.000002044197,0.000007716014,0.000006930879,0.9944919,0.002866201,0.002233408,0.0002782365,0.000002336856],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1359908,0.0003715973,0.8557995,0.0002543009,0.00004060871,0.00006552156,0.00007080415,0.002898072,0.004508825],"genre_scores_gemma":[0.7641367,0.0002297482,0.2331661,0.00008356906,0.00003847395,0.00007432219,0.0002303339,0.000291656,0.001749032],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002491873,"threshold_uncertainty_score":0.005474985,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02497315419946151,"score_gpt":0.2600890440625006,"score_spread":0.2351158898630391,"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."}}