Cargando…
The Fermi–Dirac distribution provides a calibrated probabilistic output for binary classifiers
Binary classification is one of the central problems in machine-learning research and, as such, investigations of its general statistical properties are of interest. We studied the ranking statistics of items in binary classification problems and observed that there is a formal and surprising relati...
Autores principales: | Kim, Sung-Cheol, Arun, Adith S., Ahsen, Mehmet Eren, Vogel, Robert, Stolovitzky, Gustavo |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
National Academy of Sciences
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8403970/ https://www.ncbi.nlm.nih.gov/pubmed/34413191 http://dx.doi.org/10.1073/pnas.2100761118 |
Ejemplares similares
-
Fermi-Dirac equations
por: Delbourgo, Robert, et al.
Publicado: (1991) -
Deconstructing Cross-Entropy for Probabilistic Binary Classifiers
por: Ramos, Daniel, et al.
Publicado: (2018) -
Fermi-Dirac correlations in $Zo \to ppX$
por: Kucharczyk, M, et al.
Publicado: (2004) -
Fermi-Dirac correlations in $Z^{o} ->ppX$
por: Kucharczyk, M, et al.
Publicado: (2004) -
Fermi-Dirac correlations in $Z \to ppX$ events
por: Kucharczyk, M, et al.
Publicado: (2005)