On Performance and Calibration of Natural Gradient Langevin Dynamics
Producing deep neural network (DNN) models with calibrated confidence is essential for applications in many fields, such as medical image analysis, natural language processing, and robotics. Modern neural networks have been reported to be poorly calibrated compared with those from a decade ago. The...
主要な著者: | Hanif Amal Robbani, Alhadi Bustamam, Risman Adnan, Shandar Ahmad |
---|---|
フォーマット: | 論文 |
言語: | English |
出版事項: |
IEEE
2023-01-01
|
シリーズ: | IEEE Access |
主題: | |
オンライン・アクセス: | https://ieeexplore.ieee.org/document/10131934/ |
類似資料
-
Fast Sampling of Score-Based Models With Cyclical Diffusion Sampling
著者:: Karimul Makhtidi, 等
出版事項: (2024-01-01) -
On a generalization of fractional Langevin equation with boundary conditions
著者:: Zheng Kou, 等
出版事項: (2022-01-01) -
Dynamical Sampling with Langevin Normalization Flows
著者:: Minghao Gu, 等
出版事項: (2019-11-01) -
Calibration with confidence: a principled method for panel assessment
著者:: R. S. MacKay, 等
出版事項: (2017-01-01) -
Lévy-walk-like Langevin dynamics
著者:: Xudong Wang, 等
出版事項: (2019-01-01)