{"id":"W3136181012","doi":"10.1109/access.2021.3065710","title":"A Hybrid Multi-Task Learning Approach for Optimizing Deep Reinforcement Learning Agents","year":2021,"lang":"en","type":"article","venue":"IEEE Access","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Tech University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Reinforcement learning; Computer science; Artificial intelligence; Task (project management); Machine learning; Field (mathematics); Multi-task learning; Deep learning; Engineering","routes":{"ca_aff":true,"ca_fund":true,"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","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0006184776,0.0003485556,0.0003840859,0.0001887106,0.0007125878,0.001359888,0.001862676,0.00009993956,0.00003562141],"category_scores_gemma":[0.0004392159,0.0003762812,0.0002156481,0.0004875069,0.00004622333,0.001518324,0.0008916667,0.0006478991,0.00003961982],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001627786,"about_ca_system_score_gemma":0.0001450129,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001894979,"about_ca_topic_score_gemma":8.917021e-7,"domain_scores_codex":[0.9968815,0.0001855897,0.0005941663,0.0008757195,0.0006198807,0.0008430924],"domain_scores_gemma":[0.9981209,0.0002076818,0.0004047782,0.0007150714,0.0003539771,0.0001975854],"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":[0.000009208357,0.00004408321,0.0007452704,0.000100958,0.00007889352,0.00003584619,0.0006615833,0.9927038,0.0006136471,0.0003745431,0.0002713305,0.004360869],"study_design_scores_gemma":[0.001166678,0.0001009953,0.0001052644,0.00004827808,0.00002896,0.00003182616,0.000102732,0.9843768,0.008048291,0.0000222347,0.005520173,0.0004477004],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002067663,0.000114452,0.9929256,0.00008263794,0.0007487669,0.0004996199,3.099339e-7,0.0004225491,0.003138366],"genre_scores_gemma":[0.7165096,0.00006695959,0.277349,0.0004110755,0.0001650923,0.0001457312,0.00007254886,0.00005419055,0.005225824],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7155766,"threshold_uncertainty_score":0.9998689,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0610404448592446,"score_gpt":0.314948900322499,"score_spread":0.2539084554632544,"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."}}