{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001166328,0.001637608,0.0007456832,0.00110319,0.0007876746,0.002622104,0.004980423,0.001267201,0.02789381],"category_scores_gemma":[0.009196978,0.0008167977,0.0009525234,0.001561685,0.001445922,0.007312654,0.00386247,0.003966669,0.008236932],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001606669,"about_ca_system_score_gemma":0.00250786,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003235197,"about_ca_topic_score_gemma":0.005191203,"domain_scores_codex":[0.9985679,0.0002573696,0.0001289596,0.0003152369,0.0004899818,0.0002405331],"domain_scores_gemma":[0.9967765,0.001110438,0.0001228282,0.001348655,0.000481763,0.0001598089],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003007309,0.0004768399,0.003665822,0.001101565,0.0002361526,0.0006161964,0.0006169484,0.04861915,0.06817888,0.0861802,0.1111721,0.6761289],"study_design_scores_gemma":[0.0006006329,0.0005170654,0.001459351,0.0002395082,0.0002175047,0.0007011697,0.0001758025,0.6072713,0.20109,0.08928952,0.09827826,0.0001598424],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08572274,0.003174991,0.7587298,0.003081117,0.00115152,0.0002482477,0.001201454,0.108179,0.0385111],"genre_scores_gemma":[0.5490863,0.001110733,0.4094335,0.001873294,0.0002597902,0.0004393169,0.001601076,0.009123912,0.02707222],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02789381,"threshold_uncertainty_score":0.09331411,"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."}}