{"id":"W3175853876","doi":"10.1609/aaai.v35i11.17159","title":"Online Class-Incremental Continual Learning with Adversarial Shapley Value","year":2021,"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":175,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Forgetting; Computer science; Machine learning; Class (philosophy); Artificial intelligence; Adversarial system; Memory footprint; Task (project management); Stability (learning theory); 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":[0.003223598,0.001277024,0.001890657,0.000657025,0.0005722673,0.001096338,0.003903857,0.001779827,0.002790006],"category_scores_gemma":[0.006947002,0.0005460815,0.0009361383,0.0006307428,0.001768642,0.003694095,0.002606465,0.003318869,0.0007364998],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001142589,"about_ca_system_score_gemma":0.001108253,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002326441,"about_ca_topic_score_gemma":0.002733754,"domain_scores_codex":[0.9989069,0.0003735574,0.00005809803,0.0003185795,0.0002366667,0.0001062905],"domain_scores_gemma":[0.9961711,0.002093506,0.000300519,0.0008075978,0.0004255813,0.0002016853],"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":[0.0003704806,0.0002669166,0.001673521,0.0001275583,0.0001376663,0.0001155322,0.000157607,0.7816913,0.003280417,0.01638643,0.003903548,0.191889],"study_design_scores_gemma":[0.000008773543,0.00003736278,0.00006060035,0.000004269453,0.000004894579,0.00001668746,0.000005856647,0.9922259,0.0005990404,0.006854043,0.0001763568,0.000006247254],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06644101,0.0008056259,0.9273199,0.0005594625,0.0001086137,0.0001099306,0.0001378684,0.002031913,0.002485689],"genre_scores_gemma":[0.8699688,0.0002393595,0.1235368,0.0004375814,0.0001182191,0.000142999,0.0003809903,0.000171932,0.00500335],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003903857,"threshold_uncertainty_score":0.01704824,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05678288123306853,"score_gpt":0.2831853711185195,"score_spread":0.226402489885451,"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."}}