{"id":"W4404651431","doi":"10.1007/s10601-024-09377-4","title":"Learning and fine-tuning a generic value-selection heuristic inside a constraint programming solver","year":2024,"lang":"en","type":"article","venue":"Constraints","topic":"Constraint Satisfaction and Optimization","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"Natural Sciences and Engineering Research Council of Canada; Centre interuniversitaire de recherche sur les reseaux d'entreprise, la logistique et le transport","keywords":"Solver; Heuristic; Selection (genetic algorithm); Constraint (computer-aided design); Computer science; Problem solver; Mathematical optimization; Value (mathematics); Constraint satisfaction problem; Constraint programming; Artificial intelligence; Mathematics; Machine learning; Stochastic programming","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":[],"consensus_categories":[],"category_scores_codex":[0.0003524495,0.0002048834,0.000175858,0.0002747068,0.0002664409,0.0006753871,0.0001335877,0.00009559361,0.0001462485],"category_scores_gemma":[0.0001873048,0.0002106767,0.00006766489,0.0005708089,0.0003309993,0.0004402512,0.0001039062,0.0003767916,0.00004114143],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000776888,"about_ca_system_score_gemma":0.0002615194,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001245003,"about_ca_topic_score_gemma":0.00001656604,"domain_scores_codex":[0.9984928,0.0001038508,0.0002968942,0.0005479497,0.0002145696,0.000343914],"domain_scores_gemma":[0.9993849,0.0001894911,0.00006957162,0.0001247704,0.00007826272,0.0001529803],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001996735,0.00001173047,0.00126258,0.00004981092,0.00003749821,0.00006777525,0.0006706637,0.0006676386,0.002133264,0.0421678,0.00005645417,0.9528728],"study_design_scores_gemma":[0.0004742204,0.0001528675,0.002219914,0.000288962,0.00004286103,0.001535991,0.0003415831,0.97936,0.0003741762,0.001753261,0.01292516,0.0005310091],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0146664,0.0004295225,0.9792382,0.0004842143,0.0005325223,0.0002530348,0.000002018879,0.0008859407,0.003508128],"genre_scores_gemma":[0.9170159,0.0000497589,0.08254764,0.00009158952,0.00007392181,0.00002156692,0.000006167615,0.00001671256,0.000176707],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9786924,"threshold_uncertainty_score":0.8591148,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01497864478727082,"score_gpt":0.2519402801471005,"score_spread":0.2369616353598297,"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."}}