{"id":"W4280600483","doi":"10.1016/j.aquaculture.2022.738334","title":"Using a convolutional neural network for fingerling counting: A multi-task learning approach","year":2022,"lang":"en","type":"article","venue":"Aquaculture","topic":"Water Quality Monitoring Technologies","field":"Environmental Science","cited_by":17,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Biology; Task (project management); Convolutional neural network; Artificial neural network; Artificial intelligence; Machine learning; Computer science","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":[],"consensus_categories":[],"category_scores_codex":[0.0004881639,0.0001531193,0.0001508519,0.0000194985,0.001148674,0.00005073228,0.0003554268,0.00007898307,0.00006245061],"category_scores_gemma":[0.00009903317,0.0001438571,0.00008581495,0.000272938,0.0001129979,0.0001367458,0.0005679883,0.0004723567,0.000009533096],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003033904,"about_ca_system_score_gemma":0.000009073903,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002106624,"about_ca_topic_score_gemma":0.000005736269,"domain_scores_codex":[0.9985838,0.00009343482,0.0001878799,0.0003798986,0.0003237434,0.0004312625],"domain_scores_gemma":[0.9996141,0.00005495301,0.0001279163,0.0001563693,0.00001050413,0.00003612168],"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":[0.00001966477,0.00006307253,0.04866929,0.00001235301,0.00001735942,0.000002569985,0.001001654,0.9356444,0.01124713,0.0002576027,0.002702711,0.0003622334],"study_design_scores_gemma":[0.0007827115,0.0001140309,0.006286264,0.00001130228,0.00004138266,0.00005505418,0.002418535,0.8981749,0.0004494799,0.0007731329,0.09038694,0.0005063056],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9748496,0.0002020679,0.02298616,0.000383113,0.0003750986,0.0005233117,0.00002657753,0.0004885942,0.0001654225],"genre_scores_gemma":[0.8781759,9.328782e-7,0.1207381,0.00009728082,0.0001525297,0.0001192247,0.00005713334,0.00002093485,0.0006380125],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.09775192,"threshold_uncertainty_score":0.8834789,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07641406536170983,"score_gpt":0.2872422972042587,"score_spread":0.2108282318425489,"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."}}