{"id":"W4387323330","doi":"10.48550/arxiv.2310.00229","title":"Consciousness-Inspired Spatio-Temporal Abstractions for Better Generalization in Reinforcement Learning","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"AI-based Problem Solving and Planning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"McGill University; Microsoft Research","keywords":"Abstraction; Generalization; Computer science; Artificial intelligence; Mathematics; Epistemology","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004245863,0.0002353134,0.0002465552,0.0004329555,0.0002758694,0.0001509135,0.0006634406,0.0002478749,0.00001827463],"category_scores_gemma":[0.00004991336,0.0002998949,0.0001113181,0.000428687,0.00003740028,0.0003320754,0.0004726812,0.000487608,0.00005906406],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002305055,"about_ca_system_score_gemma":0.000220547,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009391927,"about_ca_topic_score_gemma":0.00050904,"domain_scores_codex":[0.9984017,0.0001006205,0.000294529,0.0007696633,0.00008990088,0.0003435691],"domain_scores_gemma":[0.9987752,0.0001841639,0.0003279719,0.0004930189,0.000134969,0.00008465933],"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":[0.00001556839,0.00001599454,0.03609327,0.00007847338,0.00002939357,0.0000568818,0.0003213892,0.9519743,0.00001212112,0.01048512,0.0007085475,0.0002088778],"study_design_scores_gemma":[0.0004876723,0.00005261233,0.002349657,0.0001483159,0.00002450275,9.572442e-7,0.00004230123,0.9850525,0.00005330948,0.009653274,0.001787918,0.0003470098],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1252778,0.00001322361,0.8726484,0.0003780817,0.0006151872,0.0003987505,0.000009242133,0.0003876485,0.0002717183],"genre_scores_gemma":[0.9917252,0.0000281487,0.004669444,0.0001103626,0.0000823038,0.000008944898,0.0004261016,0.00002347682,0.002926015],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8679789,"threshold_uncertainty_score":0.9999453,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08887406027623773,"score_gpt":0.2178953617080649,"score_spread":0.1290213014318271,"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."}}