{"id":"W4283787033","doi":"10.1088/2632-2153/ac7ddc","title":"Curiosity in exploring chemical spaces: intrinsic rewards for molecular reinforcement learning","year":2022,"lang":"en","type":"article","venue":"Machine Learning Science and Technology","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Vector Institute; University of Toronto","funders":"Natural Resources Canada; Mitacs; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Austrian Science Fund; Compute Canada","keywords":"Curiosity; Reinforcement learning; Pace; Chemical space; Computer science; Space (punctuation); Field (mathematics); Reinforcement; Artificial intelligence; Human–computer interaction; Cognitive science; Drug discovery; Engineering; Psychology; Neuroscience; Biology; Bioinformatics; Mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.002345812,0.0001256463,0.00017961,0.0008970116,0.0006406242,0.0001165914,0.0009717948,0.00003502906,0.00000516368],"category_scores_gemma":[0.001375262,0.0001362497,0.00002792073,0.002764834,0.0002569236,0.0004256912,0.002099111,0.000751383,0.000001306427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002295169,"about_ca_system_score_gemma":0.0002045899,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004995729,"about_ca_topic_score_gemma":0.000001970793,"domain_scores_codex":[0.9981398,0.000103456,0.0002111551,0.0006111633,0.0005134871,0.0004209435],"domain_scores_gemma":[0.9993295,0.0001695841,0.0001032528,0.0002441835,0.00009953486,0.00005391347],"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":[0.00001948848,0.0000509003,0.01127161,0.00002353538,0.000007543771,0.00003636665,0.001189063,0.4926814,0.02913209,0.2725867,0.000006590775,0.1929947],"study_design_scores_gemma":[0.0005107545,0.0003467291,0.0004744153,0.00001337428,0.000003186631,0.00005796646,0.0002940754,0.9611632,0.01324786,0.01662356,0.007027898,0.0002370284],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7871503,0.0001679537,0.2082607,0.003691383,0.0001296491,0.0001984699,3.143866e-7,0.000232588,0.0001686287],"genre_scores_gemma":[0.9743515,0.00001052346,0.0252903,0.0001258929,0.000009311921,0.0001698059,0.000002326818,0.00000762912,0.00003275921],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4684817,"threshold_uncertainty_score":0.5556104,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01967815809897593,"score_gpt":0.2802007172333917,"score_spread":0.2605225591344157,"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."}}