Heuristic ML explanations for boosted trees are inadequate for the vast majority of instances on most datasets when the entire instance space is considered.
In: CExAIIA (2017)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
On Validating, Repairing and Refining Heuristic ML Explanations
Heuristic ML explanations for boosted trees are inadequate for the vast majority of instances on most datasets when the entire instance space is considered.