{"id":"W2947612752","doi":"10.48550/arxiv.1905.12588","title":"Meta-Learning Representations for Continual Learning","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Forgetting; Computer science; Artificial intelligence; Machine learning; Competitive learning; Representation (politics); Artificial neural network; Function (biology); Proactive learning; Online learning; Feature learning; Incremental learning; Robot learning; Multimedia","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001490802,0.0006288158,0.0007780912,0.0007508608,0.0004422098,0.001055257,0.002165153,0.001248339,0.003052055],"category_scores_gemma":[0.007315697,0.0005018124,0.0006052955,0.0005595055,0.001189636,0.003426579,0.002110707,0.002531834,0.0007166372],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000864359,"about_ca_system_score_gemma":0.000703079,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001175046,"about_ca_topic_score_gemma":0.001841577,"domain_scores_codex":[0.9995321,0.0001457647,0.00003110134,0.0001547786,0.00009068292,0.00004563334],"domain_scores_gemma":[0.9977715,0.0009973997,0.000192052,0.0006483906,0.0002655798,0.000125008],"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":[0.0002271733,0.000299916,0.001950375,0.0002204383,0.0001064393,0.000140954,0.0003808051,0.4977031,0.006457801,0.1153503,0.005541925,0.3716209],"study_design_scores_gemma":[0.00001047996,0.00003080612,0.0000800017,0.00001292893,0.000006908825,0.00002808902,0.00001754972,0.9506003,0.000973603,0.04723747,0.0009936936,0.000008039555],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01951676,0.0002942038,0.9773675,0.0003391695,0.00004350048,0.00003599369,0.00005698454,0.0009624798,0.001383441],"genre_scores_gemma":[0.6539961,0.0002900024,0.3403662,0.0002806204,0.0001058585,0.0002009897,0.0003314461,0.000213157,0.004215645],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003052055,"threshold_uncertainty_score":0.01021016,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1802138986397941,"score_gpt":0.2308606799815283,"score_spread":0.05064678134173414,"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."}}