{"id":"W3033044509","doi":"10.1609/aaai.v35i5.16484","title":"Combining Reinforcement Learning and Constraint Programming for Combinatorial Optimization","year":2021,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Constraint Satisfaction and Optimization","field":"Computer Science","cited_by":125,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Polytechnique Montréal","funders":"","keywords":"Combinatorial optimization; Mathematical optimization; Reinforcement learning; Constraint programming; Knapsack problem; Travelling salesman problem; Computer science; Optimization problem; Solver; Heuristics; Context (archaeology); Quadratic assignment problem; Integer programming; Lin–Kernighan heuristic; Mathematics; Stochastic programming; 2-opt; Artificial intelligence","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":[0.0003616845,0.0001406656,0.0001766624,0.0000584432,0.0003077993,0.0003552356,0.0003070614,0.0000690153,0.00003847718],"category_scores_gemma":[0.0006263028,0.0001254873,0.00006652033,0.0003537303,0.000172443,0.0003015788,0.0001862149,0.0001797525,0.000002010976],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003453405,"about_ca_system_score_gemma":0.0001297115,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004315205,"about_ca_topic_score_gemma":0.000001310013,"domain_scores_codex":[0.9987566,0.00001720846,0.0004085898,0.0003443505,0.0002602009,0.0002130037],"domain_scores_gemma":[0.9985519,0.0001149301,0.0003024382,0.0001031299,0.0008621378,0.00006543025],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002154603,0.0000386939,0.0002049109,0.00003250692,0.00001204635,2.067161e-7,0.0006026276,0.0115629,0.00304412,0.8797358,0.000007985888,0.1047366],"study_design_scores_gemma":[0.0001208014,0.0001981697,0.00002138372,0.0001632886,0.00001435483,0.00001038676,0.001119015,0.8353307,0.1353212,0.02728781,0.0002280732,0.0001848817],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004494737,0.00001650323,0.9883503,0.002113572,0.0006180364,0.0005090354,0.000001000501,0.00009588645,0.00380094],"genre_scores_gemma":[0.9408212,0.0000449069,0.05889666,0.0000740082,0.00003168352,0.00003411475,0.000002246302,0.000007429591,0.00008774458],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9363264,"threshold_uncertainty_score":0.5117223,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04400025295225241,"score_gpt":0.2862828622392445,"score_spread":0.2422826092869921,"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."}}