ConTree computes provably optimal classification trees on continuous data with new similarity-based pruning and a depth-two subroutine, running one or two orders of magnitude faster than prior optimal methods.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Optimal Classification Trees for Continuous Feature Data Using Dynamic Programming with Branch-and-Bound
ConTree computes provably optimal classification trees on continuous data with new similarity-based pruning and a depth-two subroutine, running one or two orders of magnitude faster than prior optimal methods.