{"id":"W2995911207","doi":"10.48550/arxiv.2002.01093","title":"On the interaction between supervision and self-play in emergent communication","year":2020,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Language and cultural evolution","field":"Social Sciences","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Converse; Artificial intelligence; Scratch; Machine learning; Natural language; Supervised learning; Population; Artificial neural network","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002863419,0.0004778604,0.0004549102,0.0005343996,0.0009401019,0.001126463,0.00089656,0.0009443203,0.002881056],"category_scores_gemma":[0.02336075,0.0003168976,0.0004196735,0.0002179206,0.003924085,0.003002613,0.002047878,0.001432011,0.0002391477],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001086165,"about_ca_system_score_gemma":0.0008846595,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002400638,"about_ca_topic_score_gemma":0.00238199,"domain_scores_codex":[0.9987291,0.0008530397,0.00003140944,0.0001763276,0.0001160654,0.00009411835],"domain_scores_gemma":[0.9808102,0.0158438,0.00124744,0.0008513229,0.0006317033,0.0006154701],"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":[0.000300834,0.0002946529,0.0151037,0.0002392987,0.0001207667,0.0003965483,0.002963922,0.5359268,0.01338491,0.3598768,0.002505713,0.06888612],"study_design_scores_gemma":[0.00004379704,0.0001211591,0.001499032,0.0000207255,0.00001406636,0.00007540385,0.0002574795,0.8474692,0.00166637,0.1475964,0.001209366,0.00002698767],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4124456,0.0002998337,0.5699879,0.003039365,0.00006221275,0.00008984449,0.00007115959,0.0003380886,0.01366588],"genre_scores_gemma":[0.9715082,0.0000742321,0.0270711,0.0001391618,0.00001755812,0.00005764767,0.00002347447,0.00004825747,0.00106042],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002881056,"threshold_uncertainty_score":0.01514339,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08562977860799063,"score_gpt":0.224262204123321,"score_spread":0.1386324255153303,"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."}}