{"id":"W2113839990","doi":"10.48550/arxiv.1312.6211","title":"An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networks","year":2013,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":493,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Forgetting; Dropout (neural networks); Task (project management); Computer science; Artificial neural network; Artificial intelligence; Function (biology); Machine learning; Activation function; Psychology; Cognitive psychology; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003398256,0.0002145296,0.0002658546,0.000382874,0.00008206956,0.0000913981,0.001076974,0.00020514,0.0000101306],"category_scores_gemma":[0.00003796874,0.0002545522,0.0001049728,0.0007112383,0.0001215561,0.0004896293,0.0003881855,0.0006366693,0.000006495034],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000121206,"about_ca_system_score_gemma":0.0001391221,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002514785,"about_ca_topic_score_gemma":0.00005406248,"domain_scores_codex":[0.9982248,0.0003604902,0.0002862804,0.0007209029,0.00009544891,0.0003121076],"domain_scores_gemma":[0.998497,0.0001157099,0.0003621949,0.0007126175,0.0001211774,0.000191305],"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":[0.000009116216,0.00003155619,0.09390986,0.00002474038,0.000006908737,0.00003406428,0.0003128139,0.899089,0.00003853118,0.005795892,0.00002289068,0.0007246347],"study_design_scores_gemma":[0.000380643,0.00006020578,0.04304215,0.00005565946,0.00001010315,8.822183e-7,0.00006961483,0.9529642,0.00003486155,0.003157592,0.00001100409,0.000213077],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5316223,0.00001142341,0.4678963,0.00007477219,0.0001525651,0.0001305002,0.00000120026,0.00006635714,0.00004460949],"genre_scores_gemma":[0.9929028,0.000004323527,0.006769277,0.0001934651,0.00003665277,0.000001344709,0.00005050534,0.00001289908,0.00002873672],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4612805,"threshold_uncertainty_score":0.9999906,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07776421947274287,"score_gpt":0.2129196473753434,"score_spread":0.1351554279026006,"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."}}