{"id":"W3200303296","doi":"10.1073/pnas.2106140118","title":"Deep learning for early warning signals of tipping points","year":2021,"lang":"en","type":"article","venue":"Proceedings of the National Academy of Sciences","topic":"Ecosystem dynamics and resilience","field":"Environmental Science","cited_by":274,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph; University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada; Government of Canada","keywords":"Tipping point (physics); Warning system; Computer science; False positives and false negatives; Artificial intelligence; False positive paradox; Deep learning; State (computer science); Point (geometry); Machine learning; Early warning system; Algorithm; Mathematics; Engineering","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001381667,0.00006311744,0.0001356595,0.00004743845,0.0002178444,0.00001631317,0.0005155773,0.0000505162,0.00004937456],"category_scores_gemma":[0.0007844755,0.0000446912,0.00008035079,0.0005793998,0.0004544363,0.0003072304,0.0002145798,0.0001024405,0.000001767723],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003234293,"about_ca_system_score_gemma":0.0000122515,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000920509,"about_ca_topic_score_gemma":2.897722e-7,"domain_scores_codex":[0.9985121,0.000008293584,0.0003053067,0.000216433,0.0008133117,0.0001444867],"domain_scores_gemma":[0.999204,0.0001760372,0.0004769466,0.000006278547,0.0001101107,0.00002666003],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000006250838,0.00003038879,0.1464317,0.00009625332,0.000009484417,6.849372e-9,0.000534037,0.01523217,0.8296549,0.007110677,0.00003117716,0.0008629562],"study_design_scores_gemma":[0.0002350319,0.00009334659,0.3188768,0.0003148583,0.00001547462,0.000009971015,0.0008082001,0.1330425,0.5142447,0.03202597,0.0001827661,0.0001504606],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.991401,0.00007266645,0.00006502142,0.0005983203,0.00001423439,0.0001166433,0.000002134503,0.000004552834,0.007725434],"genre_scores_gemma":[0.9959371,0.00001528734,0.003718161,0.00005052683,0.0000182976,0.00000544453,6.602299e-8,0.000003058464,0.0002520636],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3154102,"threshold_uncertainty_score":0.1822454,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02057111372301025,"score_gpt":0.2696838983739479,"score_spread":0.2491127846509376,"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."}}