{"id":"W4415428256","doi":"10.3233/faia251090","title":"Skill-Enhanced Reinforcement Learning Acceleration from Heterogeneous Demonstrations","year":2025,"lang":"","type":"book-chapter","venue":"Frontiers in artificial intelligence and applications","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Carleton University","funders":"","keywords":"Reinforcement learning; Leverage (statistics); Downstream (manufacturing); Temporal difference learning; Acceleration; Supervised learning; Online machine learning; Reinforcement","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007158076,0.0009144294,0.0007138369,0.0002213809,0.00022039,0.0005124105,0.00119071,0.0006213653,0.0049071],"category_scores_gemma":[0.002320695,0.0003825269,0.0004133959,0.0002076581,0.0005097781,0.000823068,0.001510669,0.001868487,0.001149482],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004325756,"about_ca_system_score_gemma":0.0008388574,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00203743,"about_ca_topic_score_gemma":0.002419631,"domain_scores_codex":[0.9997317,0.00005593729,0.00001306045,0.00007607094,0.00008271822,0.00004044384],"domain_scores_gemma":[0.9991799,0.0004469535,0.00007805954,0.0001220572,0.0001155311,0.00005751118],"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":[0.0001525866,0.000107981,0.0006120694,0.0001307123,0.0000322941,0.0001091823,0.00006971414,0.7507566,0.014374,0.01091862,0.00383739,0.2188989],"study_design_scores_gemma":[0.000007392037,0.00003307871,0.0000737818,0.000004977339,0.000002816822,0.00001534049,0.000002231958,0.9955584,0.001537036,0.001984146,0.0007776156,0.000003152534],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01465569,0.0002550501,0.9795024,0.0001151504,0.00006867699,0.00005509117,0.00004534304,0.001284559,0.004018047],"genre_scores_gemma":[0.6794655,0.0003633872,0.3080754,0.0001955533,0.00008029457,0.0002288542,0.0002829799,0.0002950214,0.01101302],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0049071,"threshold_uncertainty_score":0.01641589,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03288550673780755,"score_gpt":0.2740642108823426,"score_spread":0.2411787041445351,"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."}}