{"id":"W6894416521","doi":"10.5555/3635637.3662926","title":"MaDi:Learning to Mask Distractions for Generalization in Visual Deep Reinforcement Learning","year":2024,"lang":"en","type":"article","venue":"TU/e Research Portal","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Generalization; Reinforcement learning; Focus (optics); Artificial neural network; Contrast (vision); Control (management); Deep learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001353404,0.001421942,0.0008520733,0.000325569,0.0003217756,0.0006328105,0.002233192,0.001119738,0.002195888],"category_scores_gemma":[0.003458687,0.0004453235,0.0004656927,0.0001909561,0.0008573013,0.001035981,0.001561088,0.002273958,0.0004508376],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001124138,"about_ca_system_score_gemma":0.001258796,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00412616,"about_ca_topic_score_gemma":0.005166385,"domain_scores_codex":[0.9996697,0.00007056948,0.00001683586,0.0000987212,0.00007951543,0.00006473917],"domain_scores_gemma":[0.9992514,0.0003207206,0.00010334,0.0001538304,0.00008953284,0.00008098591],"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":[0.0003016129,0.0002961404,0.002264408,0.0001527455,0.0001113467,0.00009601819,0.0001449467,0.6897437,0.01377821,0.008112213,0.006947666,0.2780511],"study_design_scores_gemma":[0.00002547764,0.00007633557,0.0001241341,0.000008201995,0.000007831199,0.00001612605,0.000005551955,0.9940497,0.002830026,0.002314339,0.0005370655,0.000005149254],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06578226,0.00070259,0.920484,0.0004629569,0.0001353968,0.0001526387,0.0001404232,0.008205905,0.003933807],"genre_scores_gemma":[0.7770112,0.0001550665,0.2171904,0.0004859835,0.00005010797,0.0002359679,0.0003183734,0.0003234799,0.004229381],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00412616,"threshold_uncertainty_score":0.008204281,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05700158475965024,"score_gpt":0.4050078222895636,"score_spread":0.3480062375299134,"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."}}