Layer-wise Relevance Propagation for Neural Networks with Local Renormalization Layers
read the original abstract
Layer-wise relevance propagation is a framework which allows to decompose the prediction of a deep neural network computed over a sample, e.g. an image, down to relevance scores for the single input dimensions of the sample such as subpixels of an image. While this approach can be applied directly to generalized linear mappings, product type non-linearities are not covered. This paper proposes an approach to extend layer-wise relevance propagation to neural networks with local renormalization layers, which is a very common product-type non-linearity in convolutional neural networks. We evaluate the proposed method for local renormalization layers on the CIFAR-10, Imagenet and MIT Places datasets.
This paper has not been read by Pith yet.
Forward citations
Cited by 3 Pith papers
-
From Mechanistic to Compositional Interpretability
Compositional interpretability defines explanations as commuting syntactic-semantic mapping pairs grounded in compositionality and minimum description length, with compressive refinement and a parsimony theorem guaran...
-
From Weight Perturbation to Feature Attribution for Explaining Fully Connected Neural Networks
XWP and XWP_c are novel attribution methods for FCNNs that estimate feature importance by perturbing attached weights to avoid added bias and out-of-distribution issues in occlusion approaches.
-
Frequency-Aware Model Parameter Explorer: A new attribution method for improving explainability
FAMPE is a new attribution method that applies FFT-based frequency-selective perturbations integrated with model parameter exploration to produce fine-grained feature importance maps, showing gains over AttEXplore on ...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.