{"id":"W3014436489","doi":"10.48550/arxiv.2004.01097","title":"Learning to cooperate: Emergent communication in multi-agent navigation","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Language and cultural evolution","field":"Social Sciences","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Task (project management); Property (philosophy); Protocol (science); Space (punctuation); Core (optical fiber); Artificial intelligence; State (computer science); Human–computer interaction; State space; Communications protocol; Distributed computing; Telecommunications; Engineering; Computer network; 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":[0.0003148196,0.0001337222,0.0001598141,0.00007203782,0.0003634987,0.00005677931,0.0004633648,0.0002000002,0.0001462335],"category_scores_gemma":[0.0001406596,0.0001537279,0.00007520385,0.0006065156,0.00006354505,0.0001748867,0.0004259122,0.0005262018,0.0002291573],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005319639,"about_ca_system_score_gemma":0.0001340744,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01018758,"about_ca_topic_score_gemma":0.007701227,"domain_scores_codex":[0.9986267,0.0005057544,0.0001525754,0.0004126454,0.00009140369,0.0002109306],"domain_scores_gemma":[0.9993184,0.00002386831,0.0001155023,0.0002364223,0.000150406,0.000155445],"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":[0.0001074033,0.0002196022,0.01713822,0.00009331018,0.00007240738,0.0001584056,0.1516929,0.7908707,0.001043531,0.03518109,0.001201609,0.002220879],"study_design_scores_gemma":[0.004341461,0.0004054617,0.0686537,0.002139449,0.0004523417,0.000002043054,0.3477063,0.4618996,0.0007539333,0.01069577,0.0988058,0.004144154],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9916529,0.000117801,0.004084964,0.0009926066,0.0002068068,0.0005133976,0.000005480916,0.0001311661,0.002294898],"genre_scores_gemma":[0.9958805,0.0005768619,0.0001991929,0.0001112889,0.00005505199,0.000002694718,0.0001356806,0.000009250502,0.003029504],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3289711,"threshold_uncertainty_score":0.9964037,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1333156524103679,"score_gpt":0.262040545679653,"score_spread":0.1287248932692851,"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."}}