{"id":"W4285491035","doi":"10.1038/s41524-022-00841-4","title":"Efficient and interpretable graph network representation for angle-dependent properties applied to optical spectroscopy","year":2022,"lang":"en","type":"article","venue":"npj Computational Materials","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"Laboratory Directed Research and Development; Lawrence Livermore National Laboratory; U.S. Department of Energy","keywords":"Dihedral angle; Interpretability; Graph; Artificial neural network; Encoding (memory); Representation (politics); Graph drawing; Molecular geometry; Computer science; Algorithm; Topology (electrical circuits); Biological system; Theoretical computer science; Mathematics; Physics; Combinatorics; Artificial intelligence; Molecule; Hydrogen bond; Quantum mechanics","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001742784,0.0002038476,0.0003451507,0.0001139465,0.0007064082,0.0004453325,0.0003652844,0.00003344645,0.001078136],"category_scores_gemma":[0.0001516583,0.0001915817,0.00003359107,0.0002176847,0.0001099603,0.00006973538,0.0005657577,0.00007089393,0.00007128242],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001056104,"about_ca_system_score_gemma":0.00007163299,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004225797,"about_ca_topic_score_gemma":0.000001701074,"domain_scores_codex":[0.9974694,0.0002454065,0.0005090536,0.0006752194,0.0006418581,0.0004590921],"domain_scores_gemma":[0.9991601,0.0002145585,0.0001917765,0.0002187396,0.00009707816,0.0001177848],"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":[0.00028738,0.00003136296,0.00002356381,0.00002364197,0.000003463795,9.900464e-7,0.0002432374,0.4965002,0.4976846,0.004856904,0.0003080375,0.00003660334],"study_design_scores_gemma":[0.001431097,0.00079456,0.003686205,0.00005868461,0.00004906941,0.00006720657,0.0003885209,0.08901656,0.8760759,0.02713267,0.0005049345,0.0007945942],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.910891,0.00002489403,0.08533894,0.0004998088,0.001571164,0.001162935,0.0001201331,0.0001451873,0.0002459114],"genre_scores_gemma":[0.9030372,3.287977e-7,0.09533805,0.0004737524,0.0002447529,0.0007580615,0.00004780397,0.0000289881,0.000071092],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4074836,"threshold_uncertainty_score":0.999835,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01598162496820943,"score_gpt":0.2674490796711362,"score_spread":0.2514674547029268,"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."}}