{"id":"W4382203144","doi":"10.1609/aaai.v37i8.26169","title":"ProxyBO: Accelerating Neural Architecture Search via Bayesian Optimization with Zero-Cost Proxies","year":2023,"lang":"en","type":"article","venue":"","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute; HEC Montréal","funders":"","keywords":"Computer science; Bayesian optimization; Bayesian probability; Architecture; Task (project management); Search cost; Generalization; Machine learning; Reliability (semiconductor); Artificial intelligence; Engineering; Mathematics; Power (physics)","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":[0.0003287715,0.0001331366,0.0001019635,0.0001806529,0.0002923189,0.0004449847,0.00053459,0.00004376392,0.00002621684],"category_scores_gemma":[0.00005087806,0.00009448179,0.00002104957,0.001051341,0.00002707789,0.0004834156,0.0001818802,0.0002548841,0.00004863066],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001754588,"about_ca_system_score_gemma":0.00005206104,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005869922,"about_ca_topic_score_gemma":0.00002229715,"domain_scores_codex":[0.9986458,0.0001226895,0.0001572426,0.0004341285,0.000337341,0.0003027923],"domain_scores_gemma":[0.9992167,0.000071132,0.00005759311,0.000491903,0.00008099656,0.00008163267],"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":[0.00001492335,0.00003119134,0.00617774,0.00004402359,0.00001071451,0.00001030193,0.001227686,0.7671099,0.0007761942,0.008134542,0.0005447504,0.215918],"study_design_scores_gemma":[0.0001639418,0.00007969893,0.004843115,0.00001174321,0.000001997955,0.00001962412,0.00003360689,0.9937593,0.000434989,0.000144328,0.0003699273,0.00013778],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003311118,0.000004265471,0.9877695,0.00514978,0.00005440562,0.000379537,0.000001513436,0.0009115632,0.002418328],"genre_scores_gemma":[0.7221763,0.000002684715,0.2760598,0.0001686707,0.00005936348,0.00006436929,0.0001228133,0.00001705902,0.001328923],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7188652,"threshold_uncertainty_score":0.4290998,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02590065413059057,"score_gpt":0.2697828187619987,"score_spread":0.2438821646314081,"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."}}