{"id":"W4320561281","doi":"10.48550/arxiv.2211.02201","title":"Learning Tool Morphology for Contact-Rich Manipulation Tasks with Differentiable Simulation","year":2022,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Robot Manipulation and Learning","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Stanford Bio-X; Canadian Institute for Advanced Research; National Science Foundation","keywords":"Task (project management); GRASP; Computer science; Process (computing); Robot; Artificial intelligence; Dynamical simulation; Human–computer interaction; Programming by demonstration; Reinforcement learning; Object (grammar); Software engineering; Engineering; Programming language; Systems engineering","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001339781,0.0002931026,0.000317053,0.0002640901,0.0002702503,0.00006153958,0.0002207169,0.000232506,0.0005516962],"category_scores_gemma":[0.00002854946,0.0003572958,0.0001105322,0.0002499478,0.00001803922,0.0001768861,0.0001702756,0.0007804811,0.00002127559],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003165844,"about_ca_system_score_gemma":0.0000303552,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002962213,"about_ca_topic_score_gemma":0.00001454032,"domain_scores_codex":[0.9987934,0.00009203605,0.0002084257,0.000525326,0.00008111868,0.0002997298],"domain_scores_gemma":[0.9992173,0.0001498144,0.0001645072,0.0003221807,0.000086894,0.00005928295],"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":[0.00009442183,0.00001860538,0.01592452,0.0001387188,0.0001148119,0.0000221118,0.0001683048,0.9790221,0.0001404552,0.004200314,0.00004659782,0.0001089726],"study_design_scores_gemma":[0.000811137,0.0001047858,0.01258738,0.00003105127,0.0001359657,0.00000167907,0.0001855821,0.9836165,0.00002042655,0.0005235868,0.001562029,0.0004199288],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5462928,0.00001479161,0.4520032,0.000004704546,0.0002025115,0.0003626082,0.000002965548,0.0003288475,0.000787546],"genre_scores_gemma":[0.9957176,0.00001921178,0.0003188875,0.00001492258,0.00007989584,0.000009355364,0.0005557774,0.00007806531,0.003206316],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4516843,"threshold_uncertainty_score":0.9998879,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06745675108270803,"score_gpt":0.1963197239771882,"score_spread":0.1288629728944802,"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."}}