{"id":"W2997289589","doi":"10.1609/aaai.v34i04.5955","title":"Count-Based Exploration with the Successor Representation","year":2020,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":106,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Alberta Innovates; University of Alberta; Alberta Machine Intelligence Institute; Compute Canada","keywords":"Successor cardinal; Representation (politics); Norm (philosophy); Computer science; Sample complexity; Generalization; Reinforcement learning; Similarity (geometry); Artificial intelligence; State (computer science); Algorithm; Theoretical computer science; Mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.00008799027,0.00005680668,0.0000472838,0.00001460228,0.00009578281,0.0002344111,0.0004191969,0.00001384401,0.00002600618],"category_scores_gemma":[0.00003432348,0.00003266784,0.00001424219,0.0003411269,0.00002361452,0.0006579149,0.00004542621,0.00006576136,0.00008371771],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001110167,"about_ca_system_score_gemma":0.00004112198,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001345574,"about_ca_topic_score_gemma":0.00000441076,"domain_scores_codex":[0.9993123,0.0000410282,0.00008803108,0.0001636836,0.0003032058,0.00009168953],"domain_scores_gemma":[0.9994832,0.00007655641,0.00006603973,0.0002667173,0.00007165172,0.00003578556],"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":[0.000007721067,0.000002827602,0.0006161197,0.000003493707,0.000004180695,0.00000173571,0.0006617348,0.982796,0.0001477455,0.01208512,0.003120122,0.0005532569],"study_design_scores_gemma":[0.000206308,0.0001072943,0.0003525386,0.000002393597,0.000002930861,3.565145e-7,0.0001164978,0.9921724,0.003175796,0.00002356346,0.003780897,0.0000589864],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0007648005,0.000002439921,0.9430827,0.05225492,0.0000363995,0.0001379181,6.455837e-8,0.0001675808,0.003553123],"genre_scores_gemma":[0.9672119,0.000001073615,0.02682899,0.005538013,0.00004248473,0.00001506865,0.000004810424,0.000004986191,0.0003526612],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9664471,"threshold_uncertainty_score":0.2260432,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05162394606866699,"score_gpt":0.263603992355036,"score_spread":0.211980046286369,"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."}}