Calibrated ensembles can mitigate accuracy tradeoffs under distribution shift.

Ananya Kumar Tengyu Ma Percy Liang Aditi Raghunathan

Published in: UAI (2022)

Keyphrases

high accuracy
decision trees
computational efficiency
case study
error rate
highly accurate
computational cost
probability distribution
prediction accuracy
roc curve
bayesian networks
generalization ability
random forests
multiple classifier systems
neural network ensemble