{"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":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0005463233,0.0003702988,0.0003545822,0.0001138311,0.0002976474,0.001068886,0.00418488,0.0001633478,0.00002663679],"category_scores_gemma":[0.000485576,0.0002238458,0.0001450628,0.0007660489,0.0003729519,0.0005497353,0.001093906,0.000976394,0.00006619743],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008591529,"about_ca_system_score_gemma":0.0003782121,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006937898,"about_ca_topic_score_gemma":0.00001315505,"domain_scores_codex":[0.9969819,0.00005143351,0.0006421657,0.0007949288,0.001216653,0.0003129599],"domain_scores_gemma":[0.996449,0.0001920647,0.001293879,0.0007315352,0.001254906,0.00007859176],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001657871,0.00006620988,0.0001943477,0.0001903977,0.00005223642,0.000001062804,0.003490627,0.4201067,0.004399547,0.5628826,0.0007014713,0.007748938],"study_design_scores_gemma":[0.00003129448,0.0002203018,0.00007885056,0.000389167,0.00003623695,8.545597e-7,0.0005567255,0.7813134,0.1802936,0.03669108,0.0001166521,0.0002718753],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006363585,0.000009933099,0.938601,0.04882531,0.0005403429,0.001133053,0.000004148592,0.0001760032,0.004346611],"genre_scores_gemma":[0.9930834,0.00002379989,0.005789479,0.0006592526,0.000118782,0.0001288228,0.000005418257,0.00002363758,0.0001673885],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9867198,"threshold_uncertainty_score":0.9999681,"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."}}