{"id":"W7115933686","doi":"10.1109/tmlcn.2025.3646125","title":"Explainable Multi-Agent Reinforcement Learning for Extended Reality Codec Adaptation","year":2025,"lang":"","type":"article","venue":"IEEE Transactions on Machine Learning in Communications and Networking","topic":"Image and Video Quality Assessment","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ericsson (Canada); University of Ottawa","funders":"Mitacs; Telefonaktiebolaget LM Ericsson","keywords":"Reinforcement learning; Codec; Network packet; Overhead (engineering); Packet loss; Adaptation (eye); Decomposition; Wireless network","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":["metaepi_narrow","sts"],"consensus_categories":[],"category_scores_codex":[0.002898114,0.0004542857,0.0005491492,0.0005930956,0.003495916,0.0005786632,0.001097007,0.0002277978,0.00001536311],"category_scores_gemma":[0.00004595373,0.000531161,0.0002204649,0.001208803,0.0002034967,0.0005723991,0.00008445163,0.002049668,0.000004856677],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004550102,"about_ca_system_score_gemma":0.0002952397,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001641292,"about_ca_topic_score_gemma":0.001907868,"domain_scores_codex":[0.9952485,0.001678904,0.001229561,0.000812377,0.000315304,0.0007153598],"domain_scores_gemma":[0.995757,0.001877194,0.0004730913,0.001518322,0.000247163,0.0001271928],"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":[0.00008992402,0.000402856,0.0001564278,0.0001227924,0.00008741339,0.000001349899,0.002912088,0.5935813,0.0000394381,0.001484293,0.00001649323,0.4011057],"study_design_scores_gemma":[0.002135058,0.0004332687,0.0002854023,0.0009149823,0.0001123901,0.000003580623,0.001204624,0.9600253,0.0001159843,0.0003456363,0.03399077,0.0004330785],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0001988778,0.004784733,0.9878988,0.004302946,0.0007529416,0.001202901,0.000003816479,0.0001789707,0.000676001],"genre_scores_gemma":[0.9443577,0.01562019,0.03514601,0.0004257304,0.0000456672,0.0005293711,0.00004696715,0.00003320129,0.003795158],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9527528,"threshold_uncertainty_score":0.999714,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09524532098212235,"score_gpt":0.3653373903419221,"score_spread":0.2700920693597998,"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."}}