A study on relationship between prediction uncertainty and robustness to noisy data

DOI10.1080/00207721.2023.2169059zbMATH Open1520.93539MaRDI QIDQ6164450FDOQ6164450

Authors: Xi-Zhao Wang, Han Liu 

Publication date: 4 July 2023

Published in: International Journal of Systems Science. Principles and Applications of Systems and Integration (Search for Journal in Brave)

Recommendations

• Robustness analysis tools for an uncertainty set obtained by prediction error identification
• Identification and prediction propagation of uncertainty in models with bounded noise
• Graphical evaluation of prediction capability when the number of noise variable increases in robust parameter design
• A review of robust regression modeling approaches with noise
• Data-based prediction under uncertainty: CDF-quantile distributions and info-gap robustness
• How to handle noisy labels for robust learning from uncertainty
• Robustness, fidelity and prediction-looseness of models
• Robust estimation with unknown noise statistics
• Do we have enough data? Robust reliability via uncertainty quantification

zbMATH Keywords

classification accuracyuncertaintynoisy datatemperature parametersoftmax function

Mathematics Subject Classification ID

Sensitivity (robustness) (93B35) Stochastic systems in control theory (general) (93E03)

Cites Work

• Title not available (Why is that?)
• Class noise vs. attribute noise: A quantitative study of their impacts
• Title not available (Why is that?)
• Lectures on the nearest neighbor method
• Title not available (Why is that?)

Cited In (3)

• Graphical evaluation of prediction capability when the number of noise variable increases in robust parameter design
• Advancing neural network calibration: the role of gradient decay in large-margin Softmax optimization
• An analysis on the relationship between uncertainty and misclassification rate of classifiers

This page was built for publication: A study on relationship between prediction uncertainty and robustness to noisy data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6164450)