{"id":"W4387964087","doi":"10.48550/arxiv.2310.16686","title":"Dynamics Generalisation in Reinforcement Learning via Adaptive Context-Aware Policies","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Division of Mathematical Sciences; Canadian Institute for Advanced Research","keywords":"Adapter (computing); Reinforcement learning; Computer science; Architecture; Artificial intelligence; Robot; Context (archaeology); Machine 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001280613,0.0007699578,0.0007252725,0.0003421778,0.0002532126,0.0005683581,0.001019644,0.000756811,0.001210972],"category_scores_gemma":[0.005017309,0.0003876662,0.0004187238,0.0002287347,0.001020048,0.001106987,0.001188981,0.001417197,0.0001901153],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008456531,"about_ca_system_score_gemma":0.0007800753,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005822123,"about_ca_topic_score_gemma":0.004252578,"domain_scores_codex":[0.9994814,0.0001595177,0.00003183087,0.0001636189,0.0000968022,0.00006693045],"domain_scores_gemma":[0.9989043,0.0005625737,0.0001391518,0.0001617877,0.0001718364,0.00006021535],"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":[0.00008360668,0.00006196919,0.001214816,0.00005254112,0.00004453473,0.00007840126,0.0001595471,0.9236906,0.003816336,0.006865983,0.0002685024,0.0636632],"study_design_scores_gemma":[0.000009732719,0.00003206021,0.0001463286,0.000004494799,0.000005984495,0.000009085841,0.00000520722,0.995169,0.000516945,0.003931563,0.0001661373,0.000003506447],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1011084,0.000308039,0.8949091,0.0002442329,0.0000381826,0.00008707371,0.00002825848,0.0009842643,0.002292377],"genre_scores_gemma":[0.9563347,0.0001115091,0.04204423,0.00009144207,0.00001812703,0.00008686996,0.00003387754,0.00003989166,0.001239411],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005822123,"threshold_uncertainty_score":0.01157647,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1010150129166122,"score_gpt":0.2095667935904851,"score_spread":0.1085517806738728,"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."}}