{"id":"W4225620948","doi":"10.48550/arxiv.2112.09153","title":"An Empirical Investigation of the Role of Pre-training in Lifelong Learning","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":42,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Compute Canada; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Forgetting; Computer science; Lifelong learning; Artificial intelligence; Context (archaeology); Task (project management); Machine learning; Popularity; Set (abstract data type); Variety (cybernetics); Training (meteorology); Cognitive psychology; Psychology; Engineering","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":[],"consensus_categories":[],"category_scores_codex":[0.0004813452,0.0001437075,0.0002581692,0.0002026035,0.00007756694,0.00005389248,0.001055403,0.0001742461,0.00001017146],"category_scores_gemma":[0.0001400338,0.0001490402,0.0001223505,0.0008554867,0.0001268705,0.0003520703,0.0007742401,0.0006806976,0.000001112318],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007081612,"about_ca_system_score_gemma":0.0004923931,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001495053,"about_ca_topic_score_gemma":0.00009127316,"domain_scores_codex":[0.9983076,0.0005972615,0.0002640702,0.0005236336,0.0001326955,0.0001747591],"domain_scores_gemma":[0.9986261,0.0001358127,0.0004184192,0.0005989555,0.0001375799,0.00008313839],"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":[0.000008981253,0.00002607588,0.290141,0.00003328204,0.00001794744,0.00001283842,0.01726252,0.6782266,0.001378428,0.01105197,9.768175e-7,0.001839368],"study_design_scores_gemma":[0.0001987752,0.00003342824,0.1907524,0.0001952048,0.00001326896,0.000001429028,0.003767641,0.7959455,0.001542286,0.007344916,0.00004754314,0.0001575615],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8754112,0.00003834447,0.1234037,0.00004460351,0.0001269135,0.0001094519,6.085196e-7,0.0000451634,0.0008200392],"genre_scores_gemma":[0.9969553,0.00001595173,0.002847955,0.00004559075,0.00001922285,4.478428e-7,0.000006559415,0.000008747526,0.0001002277],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1215441,"threshold_uncertainty_score":0.6077684,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07271094776994619,"score_gpt":0.2187424208096863,"score_spread":0.1460314730397401,"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."}}