Lumberjack introduces a novel (ε,δ)-DP heavy hitter detection algorithm for hierarchical data with O(√log h) error that allows deeper trees and yields a new state-of-the-art privacy-utility tradeoff for random forests.
[2021] deterministically stop splitting when all samples in a node belong to the same class (Algorithms 1, 2, and 3, see termination condition)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Lumberjack: Better Differentially Private Random Forests through Heavy Hitter Detection in Trees
Lumberjack introduces a novel (ε,δ)-DP heavy hitter detection algorithm for hierarchical data with O(√log h) error that allows deeper trees and yields a new state-of-the-art privacy-utility tradeoff for random forests.