{"id":"W2973163720","doi":"10.48550/arxiv.1907.10134","title":"BPPSA: Scaling Back-propagation by Parallel Scan Algorithm","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Speedup; Scalability; Computer science; Recurrent neural network; Algorithm; Context (archaeology); Massively parallel; Parallel computing; Sequence (biology); Scaling; Dependency (UML); Artificial neural network; Computation; Backpropagation; Scale (ratio); Parallel algorithm; Artificial intelligence; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0001398768,0.0003620246,0.0003316138,0.0001561037,0.0002250731,0.0001368577,0.002189122,0.0002820205,0.00003320341],"category_scores_gemma":[0.00001026026,0.0004344538,0.0001722751,0.0007431422,0.0001095259,0.0005563885,0.001896767,0.0006431841,0.0009680826],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002933769,"about_ca_system_score_gemma":0.00013304,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005806217,"about_ca_topic_score_gemma":0.000006478843,"domain_scores_codex":[0.9975931,0.0001071814,0.0002509465,0.001474855,0.0001323646,0.000441571],"domain_scores_gemma":[0.9974878,0.0001145087,0.0003656856,0.001683159,0.0001613801,0.0001874007],"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":[0.00001233669,0.00008767661,0.0002685038,0.00004555149,0.00005453796,0.000031327,0.00006928322,0.9207901,0.0001257197,0.05625609,0.005886621,0.0163722],"study_design_scores_gemma":[0.000334806,0.00001702109,0.000143635,0.00005316126,0.00002612949,0.000003799769,0.00001370648,0.9610727,0.0002137256,0.03416874,0.003455752,0.000496863],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008040557,0.0001189349,0.9882143,0.0002527412,0.0004847926,0.0006284233,0.00003052949,0.000325269,0.001904475],"genre_scores_gemma":[0.9103022,0.0004422656,0.08287107,0.000262519,0.0001576764,0.000005949049,0.0001439356,0.00004393853,0.005770507],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9053432,"threshold_uncertainty_score":0.9998107,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04962296866302508,"score_gpt":0.1965533259730458,"score_spread":0.1469303573100207,"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."}}