{"id":"W4321189487","doi":"10.1016/j.patcog.2023.109428","title":"Spatio-temporal hard attention learning for skeleton-based activity recognition","year":2023,"lang":"en","type":"article","venue":"Pattern Recognition","topic":"Human Pose and Action Recognition","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Activity recognition; Computer science; Key (lock); Artificial intelligence; Benchmark (surveying); RGB color model; Skeleton (computer programming); Pattern recognition (psychology); Hidden Markov model; Reinforcement learning; Machine learning; Computer vision; Geography","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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0006813107,0.0002518093,0.0002267881,0.0005265068,0.0004758009,0.0003058173,0.0002260072,0.0001665236,0.0001936671],"category_scores_gemma":[0.0001322833,0.0002864511,0.0002414659,0.0006295285,0.00002937569,0.001126815,0.00005322312,0.0002726931,0.002923885],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001017792,"about_ca_system_score_gemma":0.00005940364,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000554718,"about_ca_topic_score_gemma":0.00006337816,"domain_scores_codex":[0.9979476,0.0002301347,0.0003176059,0.0006717407,0.0003692136,0.0004636862],"domain_scores_gemma":[0.9986854,0.0002708362,0.0003192661,0.0002479337,0.0003566469,0.0001199061],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004018077,0.0001029125,0.003242597,0.0001211521,0.00002438562,0.000007013096,0.00007540888,0.00005720213,0.002854633,0.000003460062,0.001648223,0.9918228],"study_design_scores_gemma":[0.005346885,0.001154384,0.09900171,0.0006973534,0.0001409948,0.00002833373,0.0001446597,0.7927337,0.07414608,0.0139907,0.01077257,0.001842588],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5074585,0.000002793439,0.4895688,0.0007864016,0.0005861828,0.0004761671,0.00005185692,0.0008524023,0.0002169262],"genre_scores_gemma":[0.9904171,0.00002035536,0.003514351,0.0004141554,0.0004155145,0.0005296399,0.004358859,0.00004153652,0.0002884441],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9899802,"threshold_uncertainty_score":0.9999588,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06843111957499787,"score_gpt":0.2821789431137163,"score_spread":0.2137478235387184,"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."}}