{"id":"W4240713108","doi":"10.26434/chemrxiv.13383266.v2","title":"Beyond Generative Models: Superfast Traversal, Optimization, Novelty, Exploration and Discovery (STONED) Algorithm for Molecules using SELFIES","year":2021,"lang":"en","type":"preprint","venue":"ChemRxiv","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Vector Institute; University of Toronto","funders":"Natural Resources Canada; Natural Sciences and Engineering Research Council of Canada; Austrian Science Fund; Compute Canada; École de technologie supérieure; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung; National Science Foundation","keywords":"Chemical space; Computer science; Generative grammar; Generative model; Benchmark (surveying); Deep learning; Tree traversal; Artificial intelligence; Interpolation (computer graphics); Machine learning; Virtual screening; Algorithm; Theoretical computer science; Drug discovery; Chemistry","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":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0007150655,0.0004762106,0.0005845322,0.0001256919,0.0004179808,0.001889041,0.0004491476,0.000296866,0.0001478409],"category_scores_gemma":[0.0002187822,0.0004752383,0.0001062953,0.0001543282,0.0003238824,0.001685414,0.0008050944,0.0002322687,0.00000385272],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001256677,"about_ca_system_score_gemma":0.0003549745,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001257479,"about_ca_topic_score_gemma":0.00001228645,"domain_scores_codex":[0.9972117,0.0001811908,0.0005420211,0.00122261,0.0004250064,0.0004174828],"domain_scores_gemma":[0.998442,0.0001378076,0.00037444,0.0005275013,0.0004060682,0.0001121684],"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":[0.0000100484,0.00003693538,0.000004974615,0.0001394657,0.00001385754,0.000003093992,0.001648252,0.6973677,0.300254,0.000273059,0.00004404755,0.0002045657],"study_design_scores_gemma":[0.000238612,0.00002664658,0.000003272089,0.0001046057,0.00005620827,0.000007755027,0.0005461271,0.7354779,0.2604885,0.002640275,0.000009070435,0.0004010127],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2448522,0.0004280789,0.75273,0.0002078545,0.001033963,0.000504431,0.00009298508,0.00009336528,0.00005703892],"genre_scores_gemma":[0.07310242,0.0003263526,0.924975,0.0001538257,0.0004488042,0.0001891709,0.0005897704,0.00008252684,0.0001321513],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.172245,"threshold_uncertainty_score":0.9997699,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04168675759208199,"score_gpt":0.2802105068024201,"score_spread":0.2385237492103381,"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."}}