FGVis produces fine-grained, faithful visual explanations by optimizing a pixel mask and clipping gradients during backpropagation to block adversarial evidence without added hyperparameters.
How to explain individual classification decisions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Interpretable and Fine-Grained Visual Explanations for Convolutional Neural Networks
FGVis produces fine-grained, faithful visual explanations by optimizing a pixel mask and clipping gradients during backpropagation to block adversarial evidence without added hyperparameters.