{"id":"W4391423877","doi":"10.1109/ickg59574.2023.00006","title":"Efficient Low-Rank GNN Defense Against Structural Attacks","year":2023,"lang":"en","type":"article","venue":"","topic":"Advanced Graph Neural Networks","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Carleton University","funders":"","keywords":"Adjacency matrix; Computer science; Rank (graph theory); Singular value decomposition; Graph; Maxima and minima; Sparse matrix; Algorithm; Matrix (chemical analysis); Theoretical computer science; Pattern recognition (psychology); Data mining; Artificial intelligence; Mathematics; Combinatorics","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":[],"consensus_categories":[],"category_scores_codex":[0.000123914,0.0001917702,0.0001633013,0.0001543395,0.0001934034,0.0001064025,0.0008463561,0.00006413188,0.00001849237],"category_scores_gemma":[0.00002990221,0.0001547032,0.0001109086,0.00149273,0.0000652972,0.0001517079,0.0004704737,0.0001915334,0.0005227876],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002953864,"about_ca_system_score_gemma":0.00002028303,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003197838,"about_ca_topic_score_gemma":0.0000067474,"domain_scores_codex":[0.9982697,0.00004369464,0.0002287768,0.0005293304,0.0003475118,0.0005809281],"domain_scores_gemma":[0.9988715,0.0001503524,0.00005876617,0.0007035933,0.00005952899,0.0001562525],"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":[0.00001365125,0.00002276559,0.0009876692,0.00001973472,0.00002398014,0.0002844932,0.0004809893,0.854062,0.002068072,0.05534866,0.009974031,0.07671395],"study_design_scores_gemma":[0.0002950318,0.00002983461,0.004715543,0.0000145399,0.000001919794,0.00001306579,0.00001893913,0.9900492,0.000761116,0.002999328,0.000827828,0.0002736278],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.840872,0.00005593293,0.1512891,0.0008115076,0.001253579,0.0002271042,0.0000025862,0.001683784,0.003804408],"genre_scores_gemma":[0.9864491,0.000011355,0.01162969,0.001027638,0.00009276744,0.000009630828,0.000005851934,0.00001618607,0.0007578114],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1455771,"threshold_uncertainty_score":0.671955,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01293777577321197,"score_gpt":0.257019105741954,"score_spread":0.244081329968742,"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."}}