{"id":"W3148573243","doi":"10.1145/2858788.2688521","title":"On optimizing machine learning workloads via kernel fusion","year":2015,"lang":"en","type":"article","venue":"ACM SIGPLAN Notices","topic":"Parallel Computing and Optimization Techniques","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"IBM (Canada)","funders":"Nvidia","keywords":"Computer science; Kernel (algebra); Scalability; Locality; Parallel computing; Matrix multiplication; Computation; Range (aeronautics); Algorithm","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":[0.0005363111,0.0001790892,0.0001780208,0.0001562999,0.0001969712,0.0002324509,0.001311822,0.00008463556,0.00001248452],"category_scores_gemma":[0.0004703137,0.0001584108,0.00004484956,0.0003281715,0.00002589893,0.000334295,0.0005632535,0.0002867362,0.0001346564],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000413062,"about_ca_system_score_gemma":0.0000357739,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000753057,"about_ca_topic_score_gemma":0.000004973843,"domain_scores_codex":[0.9986045,0.0001446502,0.0002188191,0.0003942506,0.0003508588,0.0002869525],"domain_scores_gemma":[0.9986106,0.0003445991,0.0001631732,0.0006311318,0.00009000061,0.0001605437],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007133963,0.0001630476,0.003961464,0.00002027498,0.0000309728,0.0000682125,0.002869463,0.907451,0.0002930863,0.006462612,0.006552627,0.07205588],"study_design_scores_gemma":[0.0003699113,0.0002943965,0.0002073689,0.00006974993,0.000005999413,0.00001116257,0.00002375453,0.9913529,0.00128385,0.003514178,0.002584126,0.0002826329],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01780638,0.0003490953,0.9710273,0.0008483063,0.0003717444,0.0001134995,6.665e-7,0.001297684,0.008185295],"genre_scores_gemma":[0.6843567,0.00001836412,0.3147604,0.0004065202,0.0000706536,0.000004028832,0.000008819362,0.0000124678,0.0003620587],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6665503,"threshold_uncertainty_score":0.6459804,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03233663507408029,"score_gpt":0.2699771110886107,"score_spread":0.2376404760145304,"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."}}