{"id":"W2970803838","doi":"","title":"Lookahead Optimizer: k steps forward, 1 step back","year":2019,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":176,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Treebank; Computer science; Stochastic gradient descent; Hyperparameter; Artificial neural network; Machine translation; Computation; Artificial intelligence; Gradient descent; Algorithm; Deep neural networks; Stability (learning theory); Mathematical optimization; Machine learning; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0002201637,0.0002265779,0.0002628099,0.0001569758,0.0002115878,0.0008607017,0.0009109073,0.00009758034,0.0000174856],"category_scores_gemma":[0.00001741534,0.0002013632,0.00006575987,0.0007221171,0.00003192507,0.008142939,0.0001809876,0.0002156813,0.002007201],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008731932,"about_ca_system_score_gemma":0.00007151863,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001065117,"about_ca_topic_score_gemma":6.553132e-7,"domain_scores_codex":[0.9980749,0.00004982296,0.0007123663,0.0002892051,0.0004708741,0.0004027605],"domain_scores_gemma":[0.9983163,0.00007131373,0.0005158592,0.0006758397,0.0002898217,0.0001308481],"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":[0.00005078074,0.0000790811,0.001189586,0.001508542,0.00003368526,0.000002678359,0.002314894,0.5254335,0.0007384669,0.05159757,0.02107513,0.395976],"study_design_scores_gemma":[0.0004528568,0.00004971702,0.00008723524,0.00008270462,0.00000335713,0.00005825325,0.0001024025,0.9085711,0.0001413025,0.00010265,0.09009764,0.0002507805],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005609719,0.0002840712,0.975105,0.0007283797,0.001075949,0.0012055,0.000006607423,0.0006379221,0.01534689],"genre_scores_gemma":[0.9554285,0.00001493913,0.04007751,0.001266755,0.0001555213,0.000173432,0.00006036416,0.00002347871,0.002799504],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9498188,"threshold_uncertainty_score":0.9987699,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01359282323168757,"score_gpt":0.2465341774647059,"score_spread":0.2329413542330183,"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."}}