The authors describe and prototype a heuristic that extracts hierarchical decision trees from feedforward neural networks by tracing activation paths, without proving equivalence for unseen inputs.
On the evaluation of the symbolic knowledge extracted from black boxes,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Deriving Equivalent Symbol-Based Decision Models from Feedforward Neural Networks
The authors describe and prototype a heuristic that extracts hierarchical decision trees from feedforward neural networks by tracing activation paths, without proving equivalence for unseen inputs.