{"id":"W2889732123","doi":"10.1609/aaai.v33i01.33013582","title":"Combined Reinforcement Learning via Abstract Representations","year":2019,"lang":"en","type":"preprint","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; McGill University","funders":"","keywords":"Reinforcement learning; Computer science; Modularity (biology); Bridging (networking); Generalization; Artificial intelligence; Representation (politics); Encoding (memory); Machine learning; Theoretical computer science; Mathematics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000852802,0.0005295855,0.0005913266,0.0002991524,0.0003212537,0.0007946454,0.004906607,0.0003288576,0.0001901879],"category_scores_gemma":[0.0007886044,0.0004474313,0.0003464405,0.0005674424,0.0002741962,0.0004625468,0.003407019,0.001962822,0.0004462048],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001653837,"about_ca_system_score_gemma":0.0003130257,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001127963,"about_ca_topic_score_gemma":0.000001963713,"domain_scores_codex":[0.9957383,0.00003462342,0.001339299,0.001016826,0.001293185,0.0005777404],"domain_scores_gemma":[0.9954885,0.0002249603,0.001884697,0.001048014,0.001223956,0.000129851],"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":[0.00003972487,0.00006284066,0.0003544266,0.0001889727,0.00005980782,6.188363e-7,0.00122924,0.695423,0.003834224,0.2926874,0.0002060907,0.005913674],"study_design_scores_gemma":[0.00004619328,0.0002845113,0.0004315256,0.0007196451,0.00003418831,0.000001834422,0.0002739636,0.8684066,0.08182137,0.04739388,0.0001072347,0.000479047],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01330946,0.00001806143,0.9031839,0.003588016,0.002620637,0.001847124,0.000002524921,0.0003556055,0.07507465],"genre_scores_gemma":[0.9926826,0.00007919101,0.003484891,0.0001537638,0.00009335329,0.00006719714,0.000007644338,0.00003414442,0.003397232],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9793731,"threshold_uncertainty_score":0.9997978,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08144392931513404,"score_gpt":0.3154658059877066,"score_spread":0.2340218766725725,"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."}}