Probabilistic decision trees incorporate uncertainty as noise in three distinct phases, with soft training reducing tree size while preserving accuracy under noise but soft evaluation showing no benefit.
Design of experiments for the nips 2003 variable selection benchmark,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.AP 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Decision Tree Learning for Uncertain Clinical Measurements
Probabilistic decision trees incorporate uncertainty as noise in three distinct phases, with soft training reducing tree size while preserving accuracy under noise but soft evaluation showing no benefit.