{"id":"W4318947280","doi":"10.1088/2632-2153/acb895","title":"Supplementing recurrent neural networks with annealing to solve combinatorial optimization problems","year":2023,"lang":"en","type":"article","venue":"Machine Learning Science and Technology","topic":"Metaheuristic Optimization Algorithms Research","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Waterloo","funders":"Vector Institute","keywords":"Travelling salesman problem; Simulated annealing; Mathematical optimization; Combinatorial optimization; Computer science; Markov chain; Optimization problem; Heuristic; Artificial neural network; Extremal optimization; Scheduling (production processes); Convergence (economics); Mathematics; Algorithm; Artificial intelligence; Machine learning","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":[0.001867249,0.0001396682,0.0001677675,0.001220655,0.0008659597,0.0004362788,0.0009406648,0.00006369643,0.00001139513],"category_scores_gemma":[0.0008546002,0.000118385,0.00001100057,0.008526829,0.000293415,0.0003808078,0.0009974389,0.0004427498,0.000008253266],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004639956,"about_ca_system_score_gemma":0.00009829389,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002550816,"about_ca_topic_score_gemma":0.000004292925,"domain_scores_codex":[0.9977584,0.00004588544,0.0002233086,0.0006589329,0.0006538804,0.0006595488],"domain_scores_gemma":[0.9989417,0.00008079448,0.0001006172,0.0003215427,0.0004148658,0.0001404896],"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":[0.000002586583,0.00001216045,0.004668598,0.000007046026,0.000003879895,0.00001010959,0.0001648598,0.9358701,0.0000676568,0.01065214,0.0000443136,0.04849654],"study_design_scores_gemma":[0.0003200156,0.0003503976,0.0001148952,0.0000167728,0.00000257471,0.00001877015,0.00004835913,0.9978641,0.00008505138,0.0002265574,0.0008079491,0.0001445776],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009729776,0.00004365764,0.9825656,0.005994358,0.0003735035,0.0003654626,8.57159e-7,0.0008179776,0.0001088292],"genre_scores_gemma":[0.7105924,0.00005472029,0.2889659,0.0001012759,0.00006779846,0.00008552076,0.00001912829,0.00002310967,0.0000901259],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7008626,"threshold_uncertainty_score":0.6660349,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01265557304297382,"score_gpt":0.2716609044891043,"score_spread":0.2590053314461305,"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."}}