{"id":"W4323651448","doi":"10.48550/arxiv.2303.04075","title":"Exploiting Trust for Resilient Hypothesis Testing with Malicious Robots (evolved version)","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Air Force Office of Scientific Research; Canadian Institute for Advanced Research","keywords":"Robot; Exploit; Computer science; Adversarial system; Artificial intelligence; Communication source; Task (project management); Statistical hypothesis testing; Computer security; Machine learning; Computer network; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0006137019,0.0005354571,0.000554234,0.0004802241,0.0006451621,0.0002263189,0.002511174,0.0003210549,0.00001244122],"category_scores_gemma":[0.001173898,0.0005870776,0.0002411545,0.001190621,0.0001306024,0.0004950673,0.002896959,0.0008425407,0.0001012044],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004546087,"about_ca_system_score_gemma":0.0003227784,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003526878,"about_ca_topic_score_gemma":0.00005645611,"domain_scores_codex":[0.9966043,0.0002018827,0.0003264974,0.001882263,0.0002296825,0.0007554391],"domain_scores_gemma":[0.9953771,0.001948944,0.0005963408,0.001487416,0.0003683984,0.0002218384],"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":[0.00009170841,0.0000402828,0.007582345,0.0001589361,0.0001051932,0.0003828488,0.0002392039,0.9814816,0.00003468138,0.00858553,0.0002387929,0.001058929],"study_design_scores_gemma":[0.001041362,0.0001513538,0.00269634,0.0004845378,0.0001452709,0.000008803908,0.0003657774,0.9793833,0.00009773428,0.01473155,0.0001155903,0.0007783477],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1503451,0.00001270069,0.8462684,0.0002382575,0.0005913271,0.0006506311,0.00001652439,0.001098592,0.0007784528],"genre_scores_gemma":[0.8413575,0.00001337632,0.1563071,0.0000680906,0.0001969772,0.000009123334,0.00001523436,0.00009079292,0.001941763],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6910124,"threshold_uncertainty_score":0.999658,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1575077665701807,"score_gpt":0.2114097435912075,"score_spread":0.05390197702102678,"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."}}