{"id":"W2997709794","doi":"10.1609/aaai.v34i04.5793","title":"On the Discrepancy between the Theoretical Analysis and Practical Implementations of Compressed Communication for Distributed Deep Learning","year":2020,"lang":"en","type":"article","venue":"","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":59,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Computer science; Implementation; Quantization (signal processing); Compression (physics); Convergence (economics); Compression ratio; Rate of convergence; Data compression; Bounded function; Upper and lower bounds; Algorithm; Data compression ratio; Layer (electronics); Theoretical computer science; Artificial intelligence; Image compression; Mathematics; Telecommunications; Image processing; Engineering; Channel (broadcasting)","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":[],"consensus_categories":[],"category_scores_codex":[0.0003504139,0.00006573868,0.0001239729,0.00003355799,0.0002599969,0.00008478696,0.0003681523,0.00002058279,0.0000355454],"category_scores_gemma":[0.0009501672,0.00003596629,0.00005921442,0.0005676673,0.0001937371,0.00009311883,0.0002044921,0.0001099144,4.992236e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000007349863,"about_ca_system_score_gemma":0.00001074953,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000824741,"about_ca_topic_score_gemma":0.00000301718,"domain_scores_codex":[0.9991666,0.0002135473,0.000218946,0.0001524965,0.0001533676,0.00009500634],"domain_scores_gemma":[0.9949459,0.00444269,0.0001261837,0.0003427574,0.0001008661,0.00004161599],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000005248777,0.00001294302,0.001845512,0.000002050125,0.00009172699,3.164477e-8,0.0005091951,0.0005243181,0.000008552465,0.9964286,0.0001005307,0.0004713023],"study_design_scores_gemma":[0.0001591713,0.0001391128,0.007861538,0.000003294297,0.0001886725,3.346608e-7,0.0003001508,0.9220063,0.0006524218,0.06857007,0.00005969944,0.00005923555],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001865281,0.000006277202,0.9553685,0.04212432,0.000003350089,0.0003678665,0.00001709843,0.00007797547,0.0001693261],"genre_scores_gemma":[0.8703065,0.000004846775,0.1292831,0.0002815011,0.000004914399,0.00005592612,0.00005900299,0.000003062361,0.00000115897],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9278585,"threshold_uncertainty_score":0.1999712,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03833580488631928,"score_gpt":0.3304518789392912,"score_spread":0.2921160740529719,"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."}}