{"id":"W4395021872","doi":"10.48550/arxiv.2404.12406","title":"Lowering PyTorch's Memory Consumption for Selective Differentiation","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Parallel Computing and Optimization Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Government of Canada; Canadian Institute for Advanced Research","keywords":"Consumption (sociology); Computer science; Sociology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002058184,0.0002397401,0.0002293885,0.0003157587,0.000146883,0.0001994864,0.0007993077,0.0002381656,0.000004474341],"category_scores_gemma":[0.00002800001,0.0002860535,0.0001900145,0.0002895099,0.0000401063,0.0001663172,0.001257812,0.0004023024,0.00003423949],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002674734,"about_ca_system_score_gemma":0.0001220908,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002562735,"about_ca_topic_score_gemma":0.000008606527,"domain_scores_codex":[0.9985351,0.00006422119,0.0001713762,0.0009239532,0.00006646527,0.0002388752],"domain_scores_gemma":[0.9989068,0.0001086268,0.0001782001,0.0005403655,0.0001916378,0.00007438593],"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":[0.00005366363,0.0001192671,0.0005515047,0.0005808639,0.0002656084,0.00004912426,0.0007191608,0.8009733,0.0001898951,0.1875199,0.005511712,0.003465956],"study_design_scores_gemma":[0.0001959647,0.00003900542,0.0003698792,0.0001355259,0.00005503303,0.00000175952,0.000004793148,0.9027282,0.002074231,0.09402897,0.00005815309,0.0003085123],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05767816,0.0001031835,0.9387513,0.00005445875,0.0008396177,0.0004391684,0.00001339033,0.001231344,0.0008894039],"genre_scores_gemma":[0.986457,0.00009357857,0.01211865,0.00003520308,0.00009182038,0.000004684043,0.00003573492,0.00002037715,0.001142876],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9287789,"threshold_uncertainty_score":0.9999592,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0639449993031951,"score_gpt":0.2147804078078019,"score_spread":0.1508354085046069,"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."}}