{"id":"W3116512656","doi":"10.1109/ictai50040.2020.00046","title":"Why Layer-Wise Learning is Hard to Scale-up and a Possible Solution via Accelerated Downsampling","year":2020,"lang":"en","type":"article","venue":"","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"Nvidia; National Aeronautics and Space Administration","keywords":"Upsampling; Computer science; Layer (electronics); Feature (linguistics); Artificial intelligence; Convolution (computer science); Feature learning; Deep learning; Scale (ratio); Backpropagation; Machine learning; Artificial neural network; 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":[0.002270489,0.00110606,0.0007387213,0.000400146,0.0007426642,0.001329592,0.001575642,0.001827615,0.004251204],"category_scores_gemma":[0.01380259,0.0007910305,0.0005081868,0.0006762665,0.001486988,0.003746159,0.001350934,0.003273091,0.002276999],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007099295,"about_ca_system_score_gemma":0.001113861,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004343143,"about_ca_topic_score_gemma":0.00495715,"domain_scores_codex":[0.9988645,0.000264068,0.00006678767,0.0003584995,0.0003072341,0.0001388992],"domain_scores_gemma":[0.9962505,0.001270562,0.0002926596,0.001350078,0.0006931769,0.0001431239],"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":[0.0005562369,0.0002197139,0.005985074,0.0007243421,0.0002726685,0.0007267736,0.000564713,0.230924,0.05985534,0.07510652,0.04755897,0.5775056],"study_design_scores_gemma":[0.00008129826,0.0001530967,0.002720875,0.0000964745,0.00006611027,0.0008039959,0.0002474809,0.8229002,0.02837841,0.1240941,0.02038252,0.00007541144],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05545462,0.003087702,0.9212022,0.01052324,0.0006520876,0.0001165892,0.0002214661,0.003610038,0.005131972],"genre_scores_gemma":[0.6212638,0.002385524,0.3609295,0.002495768,0.0006074578,0.0003273435,0.0004794757,0.0008600614,0.0106511],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004343143,"threshold_uncertainty_score":0.01422173,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07194476852431984,"score_gpt":0.2797950632922878,"score_spread":0.207850294767968,"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."}}