{"id":"W3149593300","doi":"10.1101/2021.03.28.437429","title":"Deep learning for early warning signals of regime shifts","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Ecosystem dynamics and resilience","field":"Environmental Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph; University of Waterloo","funders":"University of Waterloo","keywords":"Tipping point (physics); Autocorrelation; Warning system; Computer science; Scaling; Variance (accounting); Point (geometry); Econometrics; Regime shift; Lag; Series (stratigraphy); Statistical physics; Artificial intelligence; Algorithm; Ecology; Physics; Mathematics; Statistics; Geology; Economics; 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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009340052,0.0004415791,0.0006602145,0.0001089336,0.0002336715,0.0001754709,0.0007442459,0.0004649165,0.0001954765],"category_scores_gemma":[0.0003426165,0.0004654654,0.0002678032,0.0004166306,0.0001658253,0.0001921982,0.000940652,0.000658508,0.00004690994],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002747448,"about_ca_system_score_gemma":0.0001220855,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002230285,"about_ca_topic_score_gemma":0.00002071407,"domain_scores_codex":[0.9969684,0.0001748553,0.0006588705,0.001086431,0.0005065734,0.0006048004],"domain_scores_gemma":[0.9978309,0.0001596114,0.0007688898,0.000864889,0.0001418103,0.0002339],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.00002663711,0.0001321215,0.1182596,0.0007084548,0.0001224873,0.00004074064,0.0001201664,0.04890008,0.8315151,0.0001324771,0.0000301441,0.00001199584],"study_design_scores_gemma":[0.001001631,0.0003049427,0.6737554,0.002109637,0.0002597567,4.36307e-8,0.00005550016,0.1340278,0.1841855,0.00001080265,0.002098474,0.002190503],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9765886,0.0006809693,0.02124639,0.00004867338,0.0005229617,0.0007059238,0.00002056135,0.0001257588,0.00006014858],"genre_scores_gemma":[0.9908929,0.0001311703,0.008540381,0.00003357783,0.0001160976,0.0001571385,3.941018e-7,0.0001013751,0.00002689463],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6473296,"threshold_uncertainty_score":0.9997797,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008263118586348085,"score_gpt":0.2025972751418454,"score_spread":0.1943341565554973,"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."}}