{"id":"W3033311119","doi":"10.1109/eurosp51992.2021.00024","title":"Sponge Examples: Energy-Latency Attacks on Neural Networks","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto","funders":"Engineering and Physical Sciences Research Council; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Computer science; Energy consumption; Latency (audio); Exploit; Application-specific integrated circuit; Artificial neural network; Efficient energy use; Software portability; Energy (signal processing); Embedded system; Artificial intelligence; Computer security; Operating system; Engineering; Telecommunications","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004631058,0.0006194065,0.0006265308,0.000208655,0.0002239183,0.0008367094,0.003066922,0.0005947092,0.0003730059],"category_scores_gemma":[0.0001543,0.0005842677,0.0003332008,0.0004241036,0.00006479338,0.0003528192,0.006239044,0.002045957,0.00001970569],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000152875,"about_ca_system_score_gemma":0.0001743704,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000610276,"about_ca_topic_score_gemma":0.00007133942,"domain_scores_codex":[0.9959419,0.0004863924,0.0005482994,0.001681844,0.0006408027,0.0007007698],"domain_scores_gemma":[0.9965242,0.0004174983,0.0003301999,0.002360902,0.0001557727,0.0002114853],"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":[0.000007895635,0.00004976493,0.0004499807,0.0000224796,0.00004699876,0.0002632101,0.0002096017,0.9127043,0.000002759612,0.04057533,0.0006906966,0.04497691],"study_design_scores_gemma":[0.0002054514,0.00005063009,0.001241545,0.0001195642,0.00001707048,0.00002285811,0.00003143744,0.9951484,0.00001932146,0.001188798,0.001310171,0.0006447158],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004327658,0.0003409747,0.9724675,0.0007333608,0.005256807,0.0001507107,0.000001252799,0.0007493098,0.01597242],"genre_scores_gemma":[0.921003,0.00009132941,0.07362169,0.001762749,0.001111701,0.00004207909,0.00008432203,0.00006606308,0.002217011],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9166754,"threshold_uncertainty_score":0.9996608,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02620811630047215,"score_gpt":0.268355225118932,"score_spread":0.2421471088184599,"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."}}