{"id":"W4213397023","doi":"10.1051/0004-6361/202142751","title":"Photometric redshift estimation with convolutional neural networks and galaxy images: Case study of resolving biases in data-driven methods","year":2022,"lang":"en","type":"article","venue":"Astronomy and Astrophysics","topic":"Galaxies: Formation, Evolution, Phenomena","field":"Physics and Astronomy","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Lawrence Berkeley National Laboratory; Institut National de Physique Nucléaire et de Physique des Particules; Vanderbilt University; Institut national des sciences de l'Univers; Centre National de la Recherche Scientifique; Agence Nationale de la Recherche; York University; Office of Science; Princeton University; Alfred P. Sloan Foundation; University of Washington; College of Engineering, Michigan State University; Johns Hopkins University; Carnegie Mellon University; Harvard University; Ohio State University; National Science Foundation; New Mexico State University; University of Portsmouth; Yale University; Centre National d’Etudes Spatiales; Brookhaven National Laboratory; European Commission; U.S. Department of Energy","keywords":"Convolutional neural network; Computer science; Robustness (evolution); Photometric redshift; Artificial intelligence; Redshift; Galaxy; Machine learning; Astrophysics; Physics","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":[0.01169181,0.001094143,0.0008641895,0.001004321,0.0005169257,0.001306933,0.001829156,0.002129614,0.0004855042],"category_scores_gemma":[0.03794302,0.0004937309,0.000680529,0.001379302,0.001753638,0.002108765,0.002260029,0.002157069,0.0001821786],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001585851,"about_ca_system_score_gemma":0.001100777,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01189478,"about_ca_topic_score_gemma":0.007781887,"domain_scores_codex":[0.996899,0.001583377,0.0001633417,0.0005143474,0.000681282,0.0001587138],"domain_scores_gemma":[0.9838986,0.01060259,0.001150423,0.002274798,0.001737189,0.0003365415],"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":[0.0005590855,0.0002129183,0.08134869,0.0003618599,0.000407948,0.00049674,0.0006391226,0.7131333,0.006111601,0.03034432,0.002571031,0.1638133],"study_design_scores_gemma":[0.00003469129,0.00003590904,0.00541618,0.00004708564,0.00003040028,0.0001080126,0.0000523405,0.9658822,0.007527289,0.01909366,0.001740068,0.00003218672],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4016657,0.003242722,0.5890105,0.002260673,0.00008934052,0.0000830138,0.0003965978,0.001237579,0.002013874],"genre_scores_gemma":[0.8139997,0.0005233207,0.1833946,0.0002715654,0.0001063427,0.00004476457,0.0005286967,0.0002064842,0.0009245735],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01189478,"threshold_uncertainty_score":0.0618329,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02384274452796414,"score_gpt":0.2779764922261497,"score_spread":0.2541337476981856,"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."}}