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...
المؤلفون الرئيسيون: | Hanif Amal Robbani, Alhadi Bustamam, Risman Adnan, Shandar Ahmad |
---|---|
التنسيق: | مقال |
اللغة: | English |
منشور في: |
IEEE
2023-01-01
|
سلاسل: | IEEE Access |
الموضوعات: | |
الوصول للمادة أونلاين: | https://ieeexplore.ieee.org/document/10131934/ |
مواد مشابهة
-
Fast Sampling of Score-Based Models With Cyclical Diffusion Sampling
حسب: Karimul Makhtidi, وآخرون
منشور في: (2024-01-01) -
On a generalization of fractional Langevin equation with boundary conditions
حسب: Zheng Kou, وآخرون
منشور في: (2022-01-01) -
Dynamical Sampling with Langevin Normalization Flows
حسب: Minghao Gu, وآخرون
منشور في: (2019-11-01) -
Calibration with confidence: a principled method for panel assessment
حسب: R. S. MacKay, وآخرون
منشور في: (2017-01-01) -
Lévy-walk-like Langevin dynamics
حسب: Xudong Wang, وآخرون
منشور في: (2019-01-01)