{"id":"W3177576284","doi":"10.22323/1.395.1065","title":"Combining Maximum-Likelihood with Deep Learning for Event Reconstruction in IceCube","year":2021,"lang":"en","type":"article","venue":"Proceedings of 37th International Cosmic Ray Conference — PoS(ICRC2021)","topic":"Astrophysics and Cosmic Phenomena","field":"Physics and Astronomy","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Office of Experimental Program to Stimulate Competitive Research; Deutsches Elektronen-Synchrotron; College of Engineering, Michigan State University; Helmholtz Alliance for Astroparticle Physics; RWTH Aachen University; Vetenskapsrådet; Knut och Alice Wallenbergs Stiftelse; Fonds Wetenschappelijk Onderzoek; Deutsche Forschungsgemeinschaft; Belgian Federal Science Policy Office; Bundesministerium für Bildung und Forschung; Office of Polar Programs; Fonds De La Recherche Scientifique - FNRS; Polarforskningssekretariatet; Science and Technology Facilities Council; Michigan State University; Marquette University; University of Wisconsin-Madison; U.S. Department of Energy; National Science Foundation","keywords":"Computer science; Artificial intelligence; Event (particle physics); Context (archaeology); Machine learning; Field (mathematics); Deep learning; Process (computing); Domain (mathematical analysis); Artificial neural network; Domain knowledge; Physics; Mathematics; Astrophysics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001847028,0.000868419,0.001074975,0.0007197366,0.0005217979,0.001343891,0.001622712,0.00122656,0.002232097],"category_scores_gemma":[0.005114821,0.0007446663,0.0007243258,0.001007538,0.0008594053,0.002015761,0.001880886,0.002037241,0.0007167844],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00133446,"about_ca_system_score_gemma":0.001390414,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008782857,"about_ca_topic_score_gemma":0.01113492,"domain_scores_codex":[0.9996099,0.0001359831,0.00001528866,0.00008863233,0.00009949645,0.00005074876],"domain_scores_gemma":[0.9992331,0.0004389888,0.00007118403,0.0001072192,0.000097025,0.00005240164],"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":[0.0003798341,0.00009236301,0.003114592,0.00007201351,0.0001226918,0.0002940364,0.0001398976,0.8680428,0.001958751,0.01676606,0.005740797,0.1032762],"study_design_scores_gemma":[0.00001128967,0.000008235819,0.0001940898,0.000004417701,0.000004098115,0.00002116997,0.00001116757,0.9877619,0.0007124303,0.01062074,0.0006439661,0.000006541155],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06415464,0.0007457158,0.9191179,0.001453515,0.0001046857,0.00008562496,0.000644622,0.006176293,0.007517125],"genre_scores_gemma":[0.6156762,0.0003683756,0.3737427,0.0005329633,0.0001391285,0.0001181162,0.003144693,0.0006878563,0.005589902],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008782857,"threshold_uncertainty_score":0.01746351,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01058727224324855,"score_gpt":0.2334549784507266,"score_spread":0.2228677062074781,"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."}}