Ads
Ads
:^(
冇時間睇哂,只係睇abstract :^(
Rather, we make predictions based on the entire set of plausible models,
with contributions of models weighted by the models’ predictive value. -> 即係ensemble learning? 好似見到有AIC, 係唔係用AIC 去weight? Hierarchical model 同ensemble 有d分別,但係ensemble 都有Bayesian 版
點用Bayesian 做GLM -> 咪又係寫低個likelihood做MLE / 用MCMC去evaluate 個posterior
Ads