{"id":"W3043802286","doi":"10.1109/isca45697.2020.00075","title":"JPEG-ACT: Accelerating Deep Learning via Transform-based Lossy Compression","year":2020,"lang":"en","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":48,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"JPEG; Computer science; Lossy compression; Convolutional neural network; Data compression; Lossless JPEG; Reduction (mathematics); Image compression; Deep learning; Quantization (signal processing); Compression ratio; Artificial intelligence; JPEG 2000; Computer engineering; Computer hardware; Computer vision; Image processing; Image (mathematics)","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":[0.0002796479,0.0005996738,0.0002132896,0.0004390573,0.0001627951,0.0005152296,0.0007788471,0.0003722805,0.00464602],"category_scores_gemma":[0.00137134,0.0001584607,0.0002113886,0.0004832089,0.0003039623,0.0009186394,0.0005674993,0.0009545653,0.001308573],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004204419,"about_ca_system_score_gemma":0.0006260927,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002490761,"about_ca_topic_score_gemma":0.004619474,"domain_scores_codex":[0.9998115,0.00001292802,0.00000889435,0.00002609681,0.0001194203,0.00002110556],"domain_scores_gemma":[0.9997364,0.00008966501,0.00002189207,0.00005853044,0.00007407901,0.0000193362],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005210359,0.0002531371,0.002200289,0.0002067484,0.00006721413,0.0002881599,0.0001396798,0.131393,0.1046768,0.01789239,0.03341023,0.7089513],"study_design_scores_gemma":[0.00006299106,0.0001319119,0.0006904308,0.00002030163,0.00001285152,0.0001758137,0.00001999737,0.9060858,0.07769242,0.004652395,0.01043928,0.00001576494],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07380664,0.0007692532,0.9011676,0.0004402568,0.0002676398,0.000110517,0.0005026543,0.01354529,0.009390148],"genre_scores_gemma":[0.5288226,0.0006669189,0.4498276,0.0004147275,0.00009765832,0.0001888541,0.001897323,0.001042389,0.01704195],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00464602,"threshold_uncertainty_score":0.01554251,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03450381076356811,"score_gpt":0.2696955713069403,"score_spread":0.2351917605433722,"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."}}