{"id":"W3041603413","doi":"10.1021/acscentsci.0c00426","title":"Generative Adversarial Networks for Crystal Structure Prediction","year":2020,"lang":"en","type":"article","venue":"ACS Central Science","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":262,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto; Canadian Institute for Advanced Research","funders":"National Research Foundation of Korea","keywords":"Generative grammar; Computer science; Chemical space; Representation (politics); Crystal structure prediction; Crystal (programming language); Ternary operation; Adversarial system; Generative model; Stability (learning theory); Generative adversarial network; Space (punctuation); Artificial intelligence; Machine learning; Deep learning; Crystal structure; Drug discovery; Chemistry","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":[0.0006520956,0.0001933992,0.0001975046,0.00003926536,0.0007425994,0.0004576711,0.00102249,0.00007659086,0.0005537029],"category_scores_gemma":[0.0008264668,0.0001608999,0.00004535223,0.0005847132,0.0007539273,0.0009366789,0.0002353991,0.0001456816,0.00002216505],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001109881,"about_ca_system_score_gemma":0.0002642415,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003108081,"about_ca_topic_score_gemma":0.000007273257,"domain_scores_codex":[0.9972359,0.00008251624,0.000297354,0.0008106668,0.0006678766,0.0009056642],"domain_scores_gemma":[0.9989782,0.00006906765,0.0001689116,0.000231913,0.0001521979,0.0003996992],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006402718,0.000008000266,0.000709598,0.000008549614,0.000001166355,0.000001244381,0.0007705566,0.1007142,0.8955806,0.001054192,0.0008889873,0.00019879],"study_design_scores_gemma":[0.0007365621,0.0003883818,0.009006102,0.00001482061,0.00001542456,0.00001148904,0.0001393601,0.5279005,0.4574769,0.000588204,0.003348357,0.0003739102],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8372099,0.00002132432,0.1569781,0.001439389,0.003253349,0.0005414076,0.0001860723,0.0001975698,0.0001729073],"genre_scores_gemma":[0.9802102,0.000003152769,0.01736325,0.0009684071,0.001385172,0.0000139918,0.00001720725,0.00001290036,0.00002570721],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4381037,"threshold_uncertainty_score":0.6561307,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01083802880132513,"score_gpt":0.2411311884319809,"score_spread":0.2302931596306558,"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."}}