{"id":"W2989651853","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":"preprint","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Computer science; Implementation; Compression (physics); Quantization (signal processing); Convergence (economics); Compression ratio; Rate of convergence; Data compression; Bounded function; Algorithm; Data compression ratio; Upper and lower bounds; Artificial intelligence; Theoretical computer science; Image compression; Mathematics; Telecommunications; Engineering; Image processing; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01673164,0.002682743,0.002258016,0.001929418,0.001785803,0.005955824,0.004768834,0.005012384,0.008181457],"category_scores_gemma":[0.1152038,0.00150694,0.001072858,0.002455191,0.00748148,0.01858958,0.006565468,0.01465191,0.00208144],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00321896,"about_ca_system_score_gemma":0.003204388,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001850181,"about_ca_topic_score_gemma":0.00171554,"domain_scores_codex":[0.9862863,0.005509253,0.0007119094,0.001721023,0.004985775,0.0007856605],"domain_scores_gemma":[0.8987637,0.08052745,0.002064086,0.01163461,0.006200543,0.0008095852],"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":[0.0004878922,0.0002780552,0.001224973,0.000958345,0.00008010151,0.0002069161,0.0004597442,0.1473298,0.003696635,0.7085417,0.0143482,0.1223877],"study_design_scores_gemma":[0.00009407841,0.0001878639,0.0004556258,0.0005692054,0.00003130481,0.0003185829,0.0002000469,0.6081081,0.005630299,0.3761659,0.008166655,0.0000723892],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02127004,0.0143152,0.923121,0.01680991,0.0007006527,0.0001270003,0.0002263074,0.0009908502,0.02243899],"genre_scores_gemma":[0.6478566,0.02120788,0.3123447,0.00558489,0.002841524,0.001085491,0.0005697557,0.001298226,0.007210969],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01673164,"threshold_uncertainty_score":0.08848637,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1140298633098911,"score_gpt":0.3676228565956723,"score_spread":0.2535929932857812,"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."}}