{"id":"W2958180913","doi":"10.1093/mnras/stz1865","title":"Optimizing sparse RFI prediction using deep learning","year":2019,"lang":"en","type":"article","venue":"Monthly Notices of the Royal Astronomical Society","topic":"Radio Astronomy Observations and Technology","field":"Physics and Astronomy","cited_by":56,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"National Research Foundation; Royal Society; Australian Research Council; University of Pennsylvania; Gordon and Betty Moore Foundation; Department of Science and Technology, Republic of the Philippines; Istituto Nazionale di Astrofisica; National Science Foundation","keywords":"Reionization; HERA; Physics; Context (archaeology); Electromagnetic interference; Interferometry; Convolutional neural network; Amplitude; Visibility; Identification (biology); Interference (communication); Algorithm; Remote sensing; Computer science; Artificial intelligence; Astrophysics; Astronomy; Optics; Redshift; Particle physics; Telecommunications; Galaxy","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":[0.0001529614,0.0001739694,0.0002652694,0.00001818594,0.0002017365,0.00003712971,0.0003726086,0.00008460314,0.0002926343],"category_scores_gemma":[0.000003952054,0.000143645,0.000395604,0.0001139713,0.0001240603,0.0001496418,0.0002157875,0.0003914782,0.00001683085],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000844703,"about_ca_system_score_gemma":0.00003613572,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003208606,"about_ca_topic_score_gemma":0.000002329531,"domain_scores_codex":[0.9988973,0.0000466544,0.0003355144,0.0002766508,0.0001217922,0.0003221065],"domain_scores_gemma":[0.9992339,0.00005619567,0.0002970819,0.0003210664,0.00004020184,0.00005154862],"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":[0.000004806754,0.0000465747,0.3678992,0.000004136041,0.00007752245,5.137848e-9,0.00009426597,0.6302262,0.00008540838,0.0002346271,0.0000172757,0.001309975],"study_design_scores_gemma":[0.0004237671,0.00004132185,0.1091671,0.00002158435,0.00006172872,4.077125e-9,0.0009300642,0.8876102,0.0006021982,0.00002634642,0.00100102,0.0001146388],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9952103,0.00003538874,0.002927593,0.00009135465,0.0002574826,0.0002440638,0.00002164656,0.00004067562,0.001171452],"genre_scores_gemma":[0.9777983,2.306094e-8,0.02171956,0.000006854319,0.0001766739,0.00000933364,0.00001661927,0.00002064279,0.0002519173],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2587321,"threshold_uncertainty_score":0.5857671,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009404114708559178,"score_gpt":0.200200299109154,"score_spread":0.1907961844005948,"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."}}