On loss functions which minimize to conditional expected values and posterior probabilities

DOI10.1109/18.243457MaRDI QIDQ4307345zbMATH OpenOpenAlexFDO

Authors John W. Miller, Rod Goodman, Padhraic Smyth

Publication date 4 October 1994

Published in IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL https://authors.library.caltech.edu/93850/

zbMATH Keywords

\(P\)-admissibilitycross entropy lossmean-square error lossnecessary and sufficient conditionsobjective functionsquared-error losstraining neural networks

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Admissibility in statistical decision theory (62C15) Statistical decision theory (62C99)

Recommendations

• Some thoughts about the design of loss functions
• Loss functions
• Are Loss Functions All the Same?
• Regular conditional probabilities and strictly proper loss functions
• The fundamental nature of the log loss function

Cited in

(7)

• Risk-sensitive loss functions for sparse multi-category classification problems
• The asymptotic optimization of pre-edited ANN classifier
• Cost-sensitive learning based on Bregman divergences
• A loss function for box-constrained inverses problems
• scientific article; zbMATH DE number 4026535 (Why is no real title available?)
• Some thoughts about the design of loss functions
• Calibration methods in imbalanced binary classification

This page was built for publication: On loss functions which minimize to conditional expected values and posterior probabilities

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4307345)