Survival regression with proper scoring rules and monotonic neural networks
We consider frequently used scoring rules for right-censored survival regression models such as time-dependent concordance, survival-CRPS, integrated Brier score and integrated binomial log-likelihood, and prove that neither of them is a proper scoring rule. This means that the true survival distrib...
Auteurs principaux: | Rindt, D, Hu, R, Steinsaltz, D, Sejdinovic, D |
---|---|
Format: | Conference item |
Langue: | English |
Publié: |
Journal of Machine Learning Research
2022
|
Documents similaires
-
A kernel- and optimal transport- based test of independence between covariates and right-censored lifetimes
par: Rindt, D, et autres
Publié: (2020) -
Consistency of permutation test s of independence usingdistance covariance, HSIC and dHSIC
par: Rindt, D, et autres
Publié: (2021) -
Proper local scoring rules
par: Parry, M, et autres
Publié: (2011) -
Proper local scoring rules on discrete sample spaces
par: Dawid, A, et autres
Publié: (2011) -
Tailored proper
scoring rules elicit decision weights
par: Arthur Carvalho
Publié: (2015-01-01)