High-fidelity surrogate explanations for neural networks often fail to recover the networks' predictive advantages over linear models in regression tasks.
Costa-juss `a
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
"Faithful to What?" On the Limits of Fidelity-Based Explanations
High-fidelity surrogate explanations for neural networks often fail to recover the networks' predictive advantages over linear models in regression tasks.