{"id":"W4293107948","doi":"10.1007/s10601-022-09327-y","title":"Learning the travelling salesperson problem requires rethinking generalization","year":2022,"lang":"en","type":"article","venue":"Constraints","topic":"Advanced Graph Neural Networks","field":"Computer Science","cited_by":85,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Generalization; Computer science; Pipeline (software); Artificial intelligence; Artificial neural network; Machine learning; Deep learning; Graph; Theoretical computer science; Mathematics","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":[0.000524291,0.0001108779,0.0000946998,0.00006179079,0.001047752,0.0001170487,0.0007477321,0.00002790903,0.00005511045],"category_scores_gemma":[0.00003440845,0.00009372955,0.00005621498,0.0005172986,0.0001543085,0.0002495194,0.0002873269,0.0005028095,0.000005218551],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005210691,"about_ca_system_score_gemma":0.00004599964,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003441044,"about_ca_topic_score_gemma":0.000002296532,"domain_scores_codex":[0.9985901,0.0002784274,0.0001933396,0.0003343824,0.0003209557,0.0002828456],"domain_scores_gemma":[0.9993632,0.0001502913,0.0001379421,0.0002730745,0.00003446001,0.0000409638],"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":[0.000007471302,0.00002276766,0.00112077,0.00001119954,0.00002361223,0.00006339452,0.01461189,0.4026646,0.005781813,0.2249688,0.0004061267,0.3503176],"study_design_scores_gemma":[0.0006056878,0.0002368774,0.0006049322,0.00006157962,0.00001537568,0.0003878188,0.001636048,0.8832058,0.001382614,0.09897213,0.01228695,0.0006041623],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04000819,0.0003143734,0.9539479,0.001825218,0.0005286228,0.0003126648,0.000001359788,0.000373249,0.002688438],"genre_scores_gemma":[0.9733128,0.00002183902,0.02572163,0.0005948363,0.00007327329,0.00003110972,0.00000588286,0.00001259546,0.0002260552],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9333046,"threshold_uncertainty_score":0.8058565,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02172918509130155,"score_gpt":0.2412136384714505,"score_spread":0.2194844533801489,"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."}}