{"id":"W2886012730","doi":"10.1609/aaai.v34i04.5955","title":"Count-Based Exploration with the Successor Representation","year":2020,"lang":"en","type":"preprint","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":17,"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); Computer science; Norm (philosophy); Generalization; Sample complexity; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001841671,0.0007172966,0.001240779,0.0007941774,0.0005020114,0.001331148,0.002296543,0.001287136,0.004764724],"category_scores_gemma":[0.008967494,0.0003972143,0.0006705508,0.0008105403,0.001455981,0.003987537,0.002790785,0.002042955,0.0006649088],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00107448,"about_ca_system_score_gemma":0.001758761,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001252603,"about_ca_topic_score_gemma":0.001776898,"domain_scores_codex":[0.9989448,0.0003816857,0.00006751131,0.0002325601,0.0002639427,0.0001095487],"domain_scores_gemma":[0.9965476,0.00207214,0.000315725,0.0005690323,0.000284643,0.0002107694],"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":[0.0004218545,0.0001798642,0.002005854,0.0002008015,0.00006755615,0.0001261427,0.0002403988,0.5298722,0.003811494,0.2378902,0.003503755,0.2216799],"study_design_scores_gemma":[0.00002721016,0.00006874969,0.0000775646,0.00001327351,0.000007869891,0.00002889322,0.000009882,0.9336309,0.000837933,0.06449454,0.0007926961,0.00001049911],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02950342,0.0002134125,0.9652595,0.0003958571,0.0000570569,0.00007327984,0.0001070841,0.000958699,0.003431711],"genre_scores_gemma":[0.6727719,0.0001548462,0.3206093,0.000233351,0.00006268047,0.0003284626,0.0002834026,0.0002134409,0.005342606],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004764724,"threshold_uncertainty_score":0.01593965,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1550835111997703,"score_gpt":0.3234339624244677,"score_spread":0.1683504512246974,"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."}}