{"id":"W4404986356","doi":"10.48550/arxiv.2411.15370","title":"Deep Policy Gradient Methods Without Batch Updates, Target Networks, or Replay Buffers","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Alliance de recherche numérique du Canada","keywords":"Computer science; Real-time computing","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":[0.0008317129,0.0007038918,0.0007006336,0.001069337,0.0002158854,0.0003844278,0.003170417,0.0005461003,0.00007659311],"category_scores_gemma":[0.0002290122,0.0006953317,0.0003571316,0.002492235,0.0002161959,0.000316073,0.005478574,0.00118897,0.00006349257],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007737631,"about_ca_system_score_gemma":0.0006440474,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004361648,"about_ca_topic_score_gemma":0.00004209099,"domain_scores_codex":[0.995776,0.0004435388,0.0004858091,0.002300989,0.0001902347,0.0008034105],"domain_scores_gemma":[0.9961541,0.0002742652,0.0004086643,0.002469301,0.0002691783,0.0004245089],"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":[0.00004080452,0.00006523359,0.0002439624,0.00008084209,0.0001789573,0.0002614467,0.0003201955,0.5818316,0.000002952841,0.4132099,0.002429623,0.001334446],"study_design_scores_gemma":[0.0002163077,0.00009547461,0.00004093843,0.00014774,0.0001032263,0.00002783582,0.00003757992,0.8244115,0.0001063114,0.1730373,0.001133534,0.0006422723],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0006724349,0.0003258076,0.9914965,0.0004863128,0.001857499,0.0007625017,0.00001524998,0.001989831,0.00239387],"genre_scores_gemma":[0.4132138,0.0004315872,0.5814562,0.0003711937,0.0002833774,0.00001406447,0.00006440743,0.00009353049,0.004071828],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4125414,"threshold_uncertainty_score":0.9995498,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05825574032702813,"score_gpt":0.253147211500646,"score_spread":0.1948914711736178,"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."}}