{"id":"W3173340069","doi":"10.1145/3447686","title":"Improving Action Recognition via Temporal and Complementary Learning","year":2021,"lang":"en","type":"article","venue":"ACM Transactions on Intelligent Systems and Technology","topic":"Human Pose and Action Recognition","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University; Vector Institute","funders":"Nvidia","keywords":"Computer science; Pooling; Representation (politics); Fuse (electrical); Artificial intelligence; Feature learning; Machine learning; Action recognition; Deep learning; Pattern recognition (psychology); Temporal database; Action (physics); Data mining; Class (philosophy)","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":[0.0001389363,0.0001384271,0.0001662233,0.0003746823,0.0004208113,0.0001351579,0.0001356813,0.00014508,0.00005852218],"category_scores_gemma":[0.00001962928,0.0001447125,0.00003415977,0.0003682737,0.00005379449,0.000325716,0.00002741401,0.0003489398,0.00004121667],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004482218,"about_ca_system_score_gemma":0.00002284892,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001754923,"about_ca_topic_score_gemma":0.00009345422,"domain_scores_codex":[0.9989308,0.00007458976,0.0002773595,0.000414091,0.0001146993,0.0001884395],"domain_scores_gemma":[0.9993554,0.00007026253,0.0001054772,0.0002835141,0.0001271504,0.00005815671],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000005652526,0.00007207115,0.0002239681,0.00005770328,0.00004463887,0.00001642919,0.00009493696,0.00002000795,0.005039272,0.0008793852,0.00001749397,0.9935284],"study_design_scores_gemma":[0.003178234,0.002893082,0.0006579711,0.0009450885,0.0003020625,0.00651797,0.01713929,0.1666114,0.640708,0.07557507,0.08307951,0.002392416],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08746366,0.0003177766,0.9100022,0.001103395,0.0005195549,0.0001710041,0.000005579092,0.0003392287,0.00007761212],"genre_scores_gemma":[0.9921821,0.0004528109,0.006923619,0.00007308477,0.00003756362,0.00005851355,0.00002306634,0.00001082001,0.0002384291],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.991136,"threshold_uncertainty_score":0.5901203,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04123009817537312,"score_gpt":0.2678034478186185,"score_spread":0.2265733496432454,"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."}}