{"id":"W2800336546","doi":"10.3389/fict.2018.00006","title":"Using Spatial Reinforcement Learning to Build Forest Wildfire Dynamics Models From Satellite Images","year":2018,"lang":"en","type":"article","venue":"Frontiers in ICT","topic":"Fire effects on ecosystems","field":"Environmental Science","cited_by":60,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"University of Waterloo; Mitacs","keywords":"Reinforcement learning; Computer science; Artificial intelligence; Markov decision process; Machine learning; Random forest; Tree (set theory); Set (abstract data type); Decision tree; Process (computing); Markov process; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"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":[0.0007921665,0.0004540962,0.0004842629,0.0005331463,0.0002280107,0.0006111815,0.001046567,0.0007784375,0.001111074],"category_scores_gemma":[0.003335328,0.0004349822,0.0005754906,0.0003220629,0.0006773417,0.0009242438,0.0005551444,0.001000908,0.0002102781],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001253257,"about_ca_system_score_gemma":0.0007854814,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01371831,"about_ca_topic_score_gemma":0.01558039,"domain_scores_codex":[0.9998264,0.000056618,0.000009535882,0.00004883053,0.00003579919,0.00002285812],"domain_scores_gemma":[0.9988102,0.000741142,0.0001499516,0.00008566355,0.0001663868,0.0000467477],"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":[0.00001181712,0.00001423971,0.0006943386,0.000004583718,0.000008647254,0.00001340131,0.0000106627,0.9940774,0.0001755103,0.0005949199,0.00006200473,0.004332361],"study_design_scores_gemma":[0.000001209016,0.000002072012,0.00003980464,6.039947e-7,5.44287e-7,0.00000121881,8.567231e-7,0.9995233,0.00006702574,0.0003450488,0.00001787133,5.716806e-7],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2488752,0.0001779612,0.7472757,0.0003074759,0.00003304146,0.00008294059,0.0002092014,0.001101363,0.001937119],"genre_scores_gemma":[0.9332774,0.00005538763,0.06512728,0.00005035382,0.00001147832,0.00007036929,0.0001886254,0.0000405841,0.001178499],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01371831,"threshold_uncertainty_score":0.02727693,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008905436714029049,"score_gpt":0.2202135890507309,"score_spread":0.2113081523367019,"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."}}