{"id":"W3024781884","doi":"10.1016/j.matt.2020.04.016","title":"Crystal Site Feature Embedding Enables Exploration of Large Chemical Spaces","year":2020,"lang":"en","type":"article","venue":"Matter","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":43,"is_retracted":false,"has_abstract":false,"ca_institutions":"Ontario Tech University; National Research Council Canada; Vector Institute; University of Ottawa; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs; University of Toronto; Samsung Advanced Institute of Technology; Samsung; Government of Ontario; Canada Foundation for Innovation; Compute Canada; Ontario Research Foundation; Ontario Centres of Excellence","keywords":"Crystal (programming language); Embedding; Atom (system on chip); Autoencoder; Materials science; Crystal structure; Band gap; Doping; Density functional theory; Representation (politics); Feature (linguistics); Computer science; Physics; Chemistry; Artificial neural network; Artificial intelligence; Computational chemistry; Crystallography; Optoelectronics","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":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0003018162,0.0001222567,0.0001944819,0.00002856182,0.00006467171,0.000152198,0.000260826,0.00006069927,0.004854887],"category_scores_gemma":[0.0001182534,0.0001023069,0.00003731691,0.000128945,0.00006862114,0.0005320266,0.0001847326,0.00009561101,0.001118188],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001208911,"about_ca_system_score_gemma":0.00001469559,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001398905,"about_ca_topic_score_gemma":0.000001340161,"domain_scores_codex":[0.9988777,0.00008472123,0.0001890901,0.0003029186,0.0002885154,0.0002570373],"domain_scores_gemma":[0.999531,0.00003804442,0.0001374534,0.0001738802,0.00004277485,0.00007679327],"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":[0.00002177928,0.00001332476,0.002313451,0.00008738587,0.000001175075,0.000003053853,0.001839738,0.0005259357,0.983656,0.00008068453,0.0114502,0.000007264834],"study_design_scores_gemma":[0.000242427,0.00004159863,0.0007985827,0.00004233358,0.000009056675,0.000004863909,0.0002328427,0.005817214,0.9835538,0.0001531875,0.008909314,0.0001947837],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9842536,0.00002161167,0.006709713,0.007799123,0.0002024881,0.00009953716,0.00003745718,0.00008439735,0.0007920627],"genre_scores_gemma":[0.9868869,8.387788e-7,0.01085442,0.00176313,0.0002073997,0.000007703765,0.00001404896,0.00001569594,0.0002498236],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006035994,"threshold_uncertainty_score":0.9996595,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01519668677294133,"score_gpt":0.2696332585937563,"score_spread":0.254436571820815,"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."}}