{"id":"W3127641366","doi":"10.26434/chemrxiv.13638347.v1","title":"Deep Generative Models Enable Navigation in Sparsely Populated Chemical Space","year":2021,"lang":"en","type":"preprint","venue":"ChemRxiv","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; Canada's Michael Smith Genome Sciences Centre; University of British Columbia","funders":"","keywords":"Chemical space; Computer science; Generative model; Generative grammar; Artificial intelligence; Benchmark (surveying); Field (mathematics); Machine learning; Space (punctuation); Quality (philosophy); Drug discovery; Mathematics; Geography; Bioinformatics; Biology","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001148613,0.0004899421,0.0007258158,0.0001244229,0.0001075961,0.0006216966,0.0009389837,0.0005492614,0.001042409],"category_scores_gemma":[0.0003664833,0.0005007487,0.0001187627,0.0003752123,0.000197289,0.0003964744,0.0012793,0.000835533,0.000139274],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003551952,"about_ca_system_score_gemma":0.0002896927,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008575036,"about_ca_topic_score_gemma":0.00004197724,"domain_scores_codex":[0.9962332,0.0003362474,0.0007011719,0.001438561,0.000620757,0.0006700553],"domain_scores_gemma":[0.998156,0.00007132889,0.000442846,0.0009178203,0.0002459276,0.0001660414],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00001521982,0.00005885211,0.0002151665,0.000141895,0.000004229251,0.00004963372,0.000992986,0.2661752,0.7318835,0.0003696988,0.00006589713,0.00002783353],"study_design_scores_gemma":[0.0002097704,0.000009026036,0.0001394342,0.0002995535,0.00001376087,0.0000145286,0.00008163283,0.296876,0.6943712,0.007520596,0.00002085676,0.0004436353],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9860317,0.0003191833,0.009870569,0.0004318675,0.001392294,0.0004190021,0.000004747209,0.0002116525,0.001319004],"genre_scores_gemma":[0.9442171,0.00002700059,0.05469862,0.0001153217,0.0002703654,0.0001178985,0.0002754572,0.00005485983,0.0002233995],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.04482805,"threshold_uncertainty_score":0.9998708,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04317779125282582,"score_gpt":0.2755820241055644,"score_spread":0.2324042328527386,"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."}}