{"id":"W3195180572","doi":"10.1109/wacv56688.2023.00529","title":"Serf: Towards better training of deep neural networks using log-Softplus ERror activation Function","year":2023,"lang":"en","type":"article","venue":"2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Activation function; Artificial intelligence; Regularization (linguistics); Artificial neural network; Monotonic function; Pattern recognition (psychology); Machine learning; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003474784,0.0004203748,0.0005549876,0.0006007724,0.0002733811,0.0001376415,0.001553114,0.0002062387,0.00004955837],"category_scores_gemma":[0.00001387845,0.000423073,0.0002320224,0.002444052,0.0001962634,0.0007573174,0.0004800168,0.0004466351,0.00007862953],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009093407,"about_ca_system_score_gemma":0.00008275498,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001715325,"about_ca_topic_score_gemma":0.0000077159,"domain_scores_codex":[0.996687,0.0001448524,0.0009983737,0.001013528,0.0006033991,0.0005528628],"domain_scores_gemma":[0.9967461,0.0003271406,0.0007219012,0.001474283,0.0005527532,0.0001778018],"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":[0.00007423772,0.0001913344,0.0000772573,0.00005172559,0.00007004789,0.000002408072,0.0005551804,0.2645114,0.01344401,0.01303393,0.001322974,0.7066655],"study_design_scores_gemma":[0.00044493,0.0003170235,0.002683637,0.0001181315,0.0000258575,0.000009674624,0.0000710559,0.9874451,0.00238458,0.005319919,0.0008213304,0.0003587236],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03669823,0.00001567818,0.9592672,0.001812761,0.0005321604,0.0009362691,0.00001473774,0.0004318802,0.0002910986],"genre_scores_gemma":[0.9472864,0.00001566142,0.05124762,0.0006075028,0.0003831551,0.0002534336,0.0000773035,0.00004655508,0.00008234041],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9105882,"threshold_uncertainty_score":0.9998221,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06425871713562263,"score_gpt":0.3228914887893151,"score_spread":0.2586327716536924,"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."}}