{"id":"W3176017220","doi":"10.48550/arxiv.2106.10318","title":"Sample Efficient Social Navigation Using Inverse Reinforcement Learning","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Evacuation and Crowd Dynamics","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Sample (material); Reinforcement learning; Reinforcement; Inverse; Computer science; Artificial intelligence; Psychology; Mathematics; Social psychology; Physics","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"],"consensus_categories":[],"category_scores_codex":[0.0001225347,0.0002191939,0.000206521,0.0001300135,0.0001973861,0.00007620547,0.0001770562,0.0002515321,0.0001284336],"category_scores_gemma":[0.00002363635,0.0003128989,0.0001570572,0.0003026096,0.00004052662,0.00008490904,0.000272443,0.000582495,0.00002042728],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006429049,"about_ca_system_score_gemma":0.00008755698,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001237299,"about_ca_topic_score_gemma":0.00002066243,"domain_scores_codex":[0.999095,0.00005566151,0.0001820821,0.0003424464,0.00009256897,0.0002321943],"domain_scores_gemma":[0.9994589,0.00003528351,0.00009912434,0.0002167868,0.0001130648,0.00007681546],"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":[0.000005663206,0.00001198865,0.0004155197,0.0001246382,0.00006718485,0.00003229597,0.0006832675,0.9933515,0.000284574,0.004920288,0.00002051125,0.00008262556],"study_design_scores_gemma":[0.0002622637,0.000006047265,0.0001458596,0.00008216468,0.00008490129,0.000001095524,0.0009340864,0.9976887,0.0001186848,0.0002359116,0.000136551,0.0003037514],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5572855,0.00001000314,0.4413754,0.000003128323,0.0002664258,0.00008678625,0.00000472905,0.0001747255,0.0007932861],"genre_scores_gemma":[0.9987144,0.00004044159,0.0005399287,0.0000182527,0.00007823655,5.038937e-7,0.0003411566,0.00003343626,0.0002336625],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4414288,"threshold_uncertainty_score":0.9999323,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06372074770903365,"score_gpt":0.1950972624375552,"score_spread":0.1313765147285215,"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."}}