{"id":"W2998655644","doi":"10.1609/aaai.v34i04.5884","title":"Residual Continual Learning","year":2020,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Ministry of Science and ICT, South Korea; National Research Foundation of Korea; National Research Foundation","keywords":"Residual; Forgetting; Computer science; Normalization (sociology); Artificial intelligence; Task (project management); Machine learning; Learning network; Convolutional neural network; Deep learning; Artificial neural network; Algorithm; Engineering","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":[0.001678649,0.0009050179,0.001098355,0.0008282512,0.0005863067,0.0008870055,0.003353178,0.00127393,0.00381993],"category_scores_gemma":[0.003816241,0.0005185221,0.0006795967,0.0006980664,0.001296699,0.00264304,0.002380062,0.001913791,0.001281296],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000657805,"about_ca_system_score_gemma":0.001165674,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00222728,"about_ca_topic_score_gemma":0.002562806,"domain_scores_codex":[0.9991028,0.0001754078,0.00004209515,0.0002974997,0.0002754901,0.0001067374],"domain_scores_gemma":[0.9985425,0.0003543232,0.0001545623,0.0004207803,0.0003995328,0.0001283213],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004235347,0.0003650777,0.001830264,0.0002203365,0.0001243111,0.0001979364,0.0002295358,0.2171605,0.01633111,0.02725667,0.008448487,0.7274122],"study_design_scores_gemma":[0.00002133731,0.0001022922,0.0001779603,0.00001135981,0.00001184193,0.00007088546,0.00001469227,0.9835665,0.003548143,0.009777268,0.002680516,0.00001724895],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01882966,0.0004794831,0.9749703,0.000235531,0.0001109053,0.00006734268,0.00007552286,0.002108153,0.003123047],"genre_scores_gemma":[0.6034182,0.0003756026,0.3836921,0.0005603565,0.0002257855,0.0002362757,0.0005750869,0.0004125589,0.01050411],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00381993,"threshold_uncertainty_score":0.012779,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1016628254990642,"score_gpt":0.2867960757735737,"score_spread":0.1851332502745094,"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."}}