{"id":"W4388532585","doi":"10.5194/hess-2023-258","title":"When ancient numerical demons meet physics-informed machine learning: adjoint-based gradients for implicit differentiable modeling","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Model Reduction and Neural Networks","field":"Physics and Astronomy","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"U.S. Department of Energy; National Oceanic and Atmospheric Administration; National Science Foundation","keywords":"Automatic differentiation; Discretization; Differentiable function; Interpretability; Computer science; Applied mathematics; Artificial neural network; Maxima and minima; Mathematical optimization; Algorithm; Artificial intelligence; Mathematics; Computation","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001393495,0.0005548207,0.0006764954,0.000114984,0.0003394408,0.0001979927,0.0003890085,0.0001700641,0.0003879977],"category_scores_gemma":[0.00001343085,0.0004891652,0.0006852729,0.0001405382,0.00003510865,0.00009766308,0.0004446524,0.0007701947,0.00004609979],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009664171,"about_ca_system_score_gemma":0.0001978893,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006269891,"about_ca_topic_score_gemma":0.00001325995,"domain_scores_codex":[0.9974058,0.00005209481,0.0005955781,0.0008413901,0.0003303611,0.0007747259],"domain_scores_gemma":[0.9986324,0.0001105725,0.0002843828,0.0005120833,0.0001653955,0.000295157],"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":[0.00007355632,0.0002160448,0.001815691,0.0001093732,0.000202045,3.770188e-7,0.0001040968,0.9770072,0.00009382062,0.006225228,0.01030006,0.00385254],"study_design_scores_gemma":[0.0008014197,0.00005852254,0.00002328401,0.0000854779,0.0001083883,2.369145e-7,0.00004879613,0.9740187,0.0004478105,0.02187043,0.002030489,0.0005064696],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03758322,0.00002639265,0.9579148,0.0004307384,0.0008883774,0.001018831,0.0002328191,0.0003528099,0.001552024],"genre_scores_gemma":[0.9849063,0.000007373604,0.005427511,0.0001391608,0.0008184419,0.0006108799,0.002385712,0.000122033,0.005582632],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9524873,"threshold_uncertainty_score":0.999756,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06157307140959203,"score_gpt":0.2944952785277758,"score_spread":0.2329222071181838,"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."}}