Cargando…
Fisher information metrics for binary classifier evaluation and training
<!--HTML-->Different evaluation metrics for binary classifiers are appropriate to different scientific domains and even to different problems within the same domain. This presentation focuses on the optimisation of event selection to minimise statistical errors in HEP parameter estimation, a p...
| Autor principal: | |
|---|---|
| Lenguaje: | eng |
| Publicado: |
2018
|
| Materias: | |
| Acceso en línea: | http://cds.cern.ch/record/2312462 |