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...
Hlavní autoři: | Hanif Amal Robbani, Alhadi Bustamam, Risman Adnan, Shandar Ahmad |
---|---|
Médium: | Článek |
Jazyk: | English |
Vydáno: |
IEEE
2023-01-01
|
Edice: | IEEE Access |
Témata: | |
On-line přístup: | https://ieeexplore.ieee.org/document/10131934/ |
Podobné jednotky
-
Fast Sampling of Score-Based Models With Cyclical Diffusion Sampling
Autor: Karimul Makhtidi, a další
Vydáno: (2024-01-01) -
On a generalization of fractional Langevin equation with boundary conditions
Autor: Zheng Kou, a další
Vydáno: (2022-01-01) -
Dynamical Sampling with Langevin Normalization Flows
Autor: Minghao Gu, a další
Vydáno: (2019-11-01) -
Calibration with confidence: a principled method for panel assessment
Autor: R. S. MacKay, a další
Vydáno: (2017-01-01) -
Lévy-walk-like Langevin dynamics
Autor: Xudong Wang, a další
Vydáno: (2019-01-01)