{"id":"W2790266074","doi":"10.1016/j.neucom.2017.11.071","title":"Bayesian optimization on graph-structured search spaces: Optimizing deep multimodal fusion architectures","year":2018,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":false,"ca_institutions":"Canadian Institute for Advanced Research; University of Guelph","funders":"Air Force Research Laboratory; Defense Advanced Research Projects Agency; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Canada Foundation for Innovation; Agence Nationale de la Recherche; Nvidia","keywords":"Computer science; Artificial intelligence; Hyperparameter; Bayesian optimization; Testbed; Machine learning; Deep learning; Hyperparameter optimization; Gesture; Graph; Fuse (electrical); Optimization problem; Pattern recognition (psychology); Theoretical computer science; Support vector machine; Algorithm","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002314612,0.001285397,0.001628264,0.0009995931,0.0005906015,0.001252432,0.001679913,0.003135742,0.003785598],"category_scores_gemma":[0.007252622,0.0009613836,0.0009246731,0.001190325,0.001532384,0.002756821,0.002281308,0.002093985,0.0006392877],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001296058,"about_ca_system_score_gemma":0.001633843,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005162772,"about_ca_topic_score_gemma":0.007449794,"domain_scores_codex":[0.9994032,0.0002914858,0.00002521129,0.0001006182,0.0001034637,0.00007601376],"domain_scores_gemma":[0.9980087,0.001443789,0.0001194032,0.00008798882,0.0002319676,0.000108133],"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":[0.00006597272,0.00004082039,0.0001995912,0.00003933618,0.00003167329,0.00002532346,0.00003705141,0.9622816,0.0004057856,0.01475514,0.0009481575,0.02116958],"study_design_scores_gemma":[0.000005566138,0.000008893433,0.00001589509,0.000003956725,0.000003075423,0.000002828704,0.000004412072,0.9932653,0.00006559661,0.006562258,0.00006006868,0.000002220159],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01720246,0.0002930353,0.9794667,0.0004045051,0.00003016887,0.00003331752,0.00005269604,0.000207802,0.002309394],"genre_scores_gemma":[0.6724886,0.0003592974,0.3200755,0.0004029255,0.00008795346,0.0003052949,0.0002497631,0.0002941071,0.005736383],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005162772,"threshold_uncertainty_score":0.01266414,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01386638003423593,"score_gpt":0.2652139794667731,"score_spread":0.2513475994325371,"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."}}