{"id":"W2781641451","doi":"10.48550/arxiv.1801.01415","title":"What have we learned from deep representations for action recognition?","year":2018,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Human Pose and Action Recognition","field":"Computer Science","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Engineering and Physical Sciences Research Council; Austrian Science Fund; Canada First Research Excellence Fund; Natural Sciences and Engineering Research Council of Canada; Österreichischen Akademie der Wissenschaften; TU Graz, Internationale Beziehungen und Mobilitätsprogramme","keywords":"Computer science; Hierarchy; Motion (physics); Artificial intelligence; Action recognition; Action (physics); Class (philosophy); Software deployment; Key (lock); Representation (politics); Range (aeronautics); Deep learning","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000160429,0.0002605348,0.0002456912,0.0003175148,0.0003842808,0.0005910047,0.0007470154,0.0003309954,0.0003415974],"category_scores_gemma":[0.00005240566,0.0003279898,0.0002798577,0.0002510386,0.00008774209,0.002011495,0.0005131462,0.000370509,0.0005912917],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000173444,"about_ca_system_score_gemma":0.0001070524,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001490631,"about_ca_topic_score_gemma":0.0003458776,"domain_scores_codex":[0.9979934,0.0001532028,0.0002355364,0.001246982,0.00009793861,0.0002730108],"domain_scores_gemma":[0.9979367,0.0002457546,0.0003566932,0.0008767418,0.0004432679,0.0001408678],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001361457,0.002639213,0.0021326,0.0009354654,0.003356986,0.0005441426,0.01400096,0.05598623,0.001696332,0.05185243,0.04844101,0.8170532],"study_design_scores_gemma":[0.0008893834,0.00008596716,0.0003533147,0.0002331297,0.0001727148,0.000003371635,0.001184836,0.3045321,0.00255408,0.685542,0.003898429,0.0005506484],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1297488,0.00005448255,0.8643891,0.000732091,0.002797276,0.0005672689,0.00007128245,0.0003588576,0.001280798],"genre_scores_gemma":[0.9833376,0.002715141,0.007397928,0.000305819,0.0008771921,0.00002041981,0.0008524643,0.00003561468,0.004457777],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8569912,"threshold_uncertainty_score":0.9999172,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.259481286929266,"score_gpt":0.2643557317179868,"score_spread":0.004874444788720733,"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."}}