{"id":"W4315881264","doi":"10.48550/arxiv.2301.04268","title":"Adversarial Online Multi-Task Reinforcement Learning","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Omega; Combinatorics; Sample complexity; Regret; Reinforcement learning; Upper and lower bounds; Lambda; Mathematics; Tilde; Discrete mathematics; Finite set; Artificial intelligence; Computer science; Physics; Statistics","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":[0.002343693,0.001395289,0.001936093,0.0003609815,0.0005578891,0.001022676,0.002723247,0.00181045,0.003133966],"category_scores_gemma":[0.0087857,0.0005212784,0.0007147922,0.0006188845,0.001693478,0.002015688,0.002094523,0.002957132,0.0005373284],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001655726,"about_ca_system_score_gemma":0.001570279,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004536705,"about_ca_topic_score_gemma":0.004295117,"domain_scores_codex":[0.9985727,0.0005420641,0.00004812525,0.0004062042,0.0002087007,0.0002221976],"domain_scores_gemma":[0.9948783,0.003498649,0.0003963299,0.0006171953,0.000237805,0.0003717324],"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":[0.0001764985,0.00009678643,0.0004733684,0.0000533345,0.00003656957,0.0000782436,0.00002849264,0.9640921,0.0004043166,0.02352097,0.001410418,0.009628837],"study_design_scores_gemma":[0.00001311275,0.0000144367,0.00004019975,0.000002234616,0.00000256823,0.000006557742,0.000002216913,0.9900959,0.0001264123,0.009518521,0.000174771,0.000003050532],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04176081,0.0004347505,0.950734,0.001282415,0.000110079,0.0001024083,0.0002547282,0.0006695397,0.00465126],"genre_scores_gemma":[0.8902212,0.0001900522,0.1028001,0.000464768,0.0001272685,0.0002281321,0.0003811728,0.00009896162,0.005488318],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004536705,"threshold_uncertainty_score":0.01239479,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09278095900279544,"score_gpt":0.2197380260531538,"score_spread":0.1269570670503584,"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."}}