{"id":"W3206990726","doi":"10.1145/3487025","title":"Adaptive Computation Reuse for Energy-Efficient Training of Deep Neural Networks","year":2021,"lang":"en","type":"article","venue":"ACM Transactions on Embedded Computing Systems","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Lakehead University","funders":"","keywords":"Computer science; Computation; Speedup; Operand; Artificial neural network; Inference; Efficient energy use; Computer engineering; Reuse; Deep neural networks; Energy (signal processing); Artificial intelligence; Parallel computing; Algorithm; Computer hardware","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002746827,0.0002650893,0.0004244329,0.0001573842,0.0004657505,0.0001067487,0.00110195,0.0001160627,0.000001313431],"category_scores_gemma":[0.00006526378,0.0002889863,0.0002135986,0.001121734,0.00006011942,0.0001588252,0.00006046999,0.0002404326,0.000001623076],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001049978,"about_ca_system_score_gemma":0.00006108553,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001224561,"about_ca_topic_score_gemma":0.000007306613,"domain_scores_codex":[0.9975401,0.000218953,0.000711274,0.0007329375,0.0003313682,0.000465367],"domain_scores_gemma":[0.9961809,0.001492856,0.000390492,0.001360383,0.0004471765,0.0001281907],"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":[0.00001593129,0.0001058466,8.310787e-7,0.00001673868,0.00004485169,0.000003852488,0.001092383,0.8759573,0.0002015071,0.008472585,0.00002395156,0.1140642],"study_design_scores_gemma":[0.000577939,0.0001456813,0.0000136324,0.00007747516,0.00002784907,0.00006867237,0.0006779703,0.9959611,0.0008519659,0.001263374,0.00007935929,0.0002549302],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002639447,0.0003010075,0.9946804,0.0002906269,0.001213576,0.0004392564,0.000009860439,0.0003397493,0.00008608535],"genre_scores_gemma":[0.7968104,0.000005013492,0.2027999,0.0001146647,0.0001247467,0.00008023239,0.00001350119,0.00002785667,0.00002368863],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.794171,"threshold_uncertainty_score":0.9999563,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03808900066656054,"score_gpt":0.281637279435358,"score_spread":0.2435482787687975,"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."}}