{"id":"W3099201768","doi":"10.1088/1475-7516/2021/04/081","title":"deep21: a deep learning method for 21 cm foreground removal","year":2021,"lang":"en","type":"article","venue":"Journal of Cosmology and Astroparticle Physics","topic":"Radio Astronomy Observations and Technology","field":"Physics and Astronomy","cited_by":62,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Convolutional neural network; Pattern recognition (psychology); Principal component analysis; Cluster analysis; Deep learning; SIGNAL (programming language); Amplitude; Phase (matter); Cosmic variance","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":[0.0002471513,0.0001216426,0.0003206615,0.00003396079,0.0001777039,0.00003516125,0.0001014316,0.00004505028,0.00002739222],"category_scores_gemma":[0.00002968638,0.0001147538,0.0001511112,0.0001444554,0.00008334954,0.0001866187,0.0000457626,0.0002883881,0.000001922949],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001487346,"about_ca_system_score_gemma":0.00007395673,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000537757,"about_ca_topic_score_gemma":0.000001184716,"domain_scores_codex":[0.9990832,0.00007443057,0.0003459416,0.0001518988,0.00007400954,0.0002705156],"domain_scores_gemma":[0.9990371,0.00020234,0.0003014439,0.000120014,0.0002586925,0.00008038126],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001356465,0.0003150707,0.1015783,0.00001334066,0.0004160497,0.00003507074,0.0002141696,0.002298755,0.02058507,0.3554004,0.0001648674,0.5188432],"study_design_scores_gemma":[0.006896053,0.001626428,0.02818877,0.00005728539,0.0006074447,0.0008259217,0.003521983,0.04182847,0.08613233,0.7963133,0.03341972,0.0005822676],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4971507,0.0001217582,0.5021657,0.0003705533,0.00006949856,0.00003940984,0.000002211281,0.000005718066,0.0000743959],"genre_scores_gemma":[0.8246022,0.000003233911,0.1748896,0.0000443684,0.0003495731,0.000007827653,0.000006737985,0.00001209121,0.00008438749],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.518261,"threshold_uncertainty_score":0.4679523,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01572572900979775,"score_gpt":0.2784561613672499,"score_spread":0.2627304323574522,"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."}}