{"id":"W4400033029","doi":"10.1109/tkde.2024.3419449","title":"Hessian Aware Low-Rank Perturbation for Order-Robust Continual Learning","year":2024,"lang":"en","type":"article","venue":"IEEE Transactions on Knowledge and Data Engineering","topic":"Geophysical Methods and Applications","field":"Engineering","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval; McGill University; Vector Institute; Western University","funders":"Natural Sciences and Engineering Research Council of Canada; National Natural Science Foundation of China","keywords":"Computer science; Hessian matrix; Perturbation (astronomy); Artificial intelligence; Mathematical optimization; Machine learning; Mathematics; Applied mathematics","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.0001162414,0.0001729239,0.0001527097,0.0001129927,0.0001214994,0.0001012017,0.0001361438,0.00007764569,0.0000183423],"category_scores_gemma":[0.00001051623,0.0001734385,0.00004470298,0.0002801877,0.000014022,0.000259745,0.000002899779,0.000287131,0.00003464973],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002509044,"about_ca_system_score_gemma":0.00001516502,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002769071,"about_ca_topic_score_gemma":0.000006235016,"domain_scores_codex":[0.9992704,0.000009455857,0.0001625904,0.000299144,0.00005545838,0.0002028933],"domain_scores_gemma":[0.9993302,0.0002855879,0.000007256469,0.0002613277,0.00003363286,0.00008200423],"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":[0.00001035338,0.00009472667,0.000001178705,0.001561094,0.0001917303,0.000002648473,0.0006211213,0.5898263,0.01882163,0.001657619,0.001156996,0.3860546],"study_design_scores_gemma":[0.0001601297,0.00003048,0.00002494612,0.0001689181,0.00005842159,0.000003537498,0.00004129504,0.9577826,0.002668547,0.00004639748,0.03880616,0.0002085085],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003246762,0.0007750922,0.993885,0.00005864646,0.0006438015,0.0002148516,0.000240593,0.0006934454,0.000241808],"genre_scores_gemma":[0.9854295,0.0001799794,0.0131647,0.000005667662,0.0002612732,0.000168022,0.0001154971,0.00007137063,0.0006039434],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9821828,"threshold_uncertainty_score":0.7072617,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02589953566920272,"score_gpt":0.2802999281194727,"score_spread":0.25440039245027,"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."}}