{"id":"W4306882122","doi":"10.48550/arxiv.2101.07241","title":"Learning by Watching: Physical Imitation of Manipulation Skills from Human Videos","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institute for Advanced Research","keywords":"Computer science; Artificial intelligence; Reinforcement learning; Task (project management); Salient; Imitation; Robot; Representation (politics); Unsupervised learning; Machine learning; Deep learning; Human–computer interaction","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":[0.0001480692,0.000247195,0.0003336097,0.0001385081,0.0002074287,0.0001193308,0.001050319,0.0001903698,0.00002054545],"category_scores_gemma":[0.00007514604,0.0003143873,0.0001967386,0.0004010854,0.00005747128,0.0003531115,0.001104939,0.0008321207,0.00002691702],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001425627,"about_ca_system_score_gemma":0.00006872289,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003004395,"about_ca_topic_score_gemma":0.00003527508,"domain_scores_codex":[0.998109,0.0002989489,0.0002426575,0.00100406,0.0001467854,0.0001985088],"domain_scores_gemma":[0.9980652,0.0002627445,0.0004986592,0.0008921139,0.0001878276,0.00009348269],"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":[0.00000490889,0.0003426339,0.02444667,0.00004678858,0.0001034626,0.0000166953,0.002874319,0.908547,0.01386581,0.04759615,0.00004909402,0.002106499],"study_design_scores_gemma":[0.0003097933,0.00003783984,0.0501167,0.00008644362,0.00006336636,4.216363e-7,0.0001205609,0.9236317,0.001759283,0.02350312,0.00004252404,0.000328268],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.576324,0.000008831929,0.4229877,0.00006382254,0.00005985665,0.0001255064,0.000006252752,0.0001343752,0.0002896995],"genre_scores_gemma":[0.9935667,0.000009610465,0.005626679,0.0000200733,0.00006361244,0.000002110894,0.0004088529,0.00002127865,0.0002810701],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.417361,"threshold_uncertainty_score":0.9999308,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0409648179073554,"score_gpt":0.2176801157903314,"score_spread":0.176715297882976,"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."}}