Understanding Neural Networks through Representation Erasure
Interpretable Recurrent Neural Networks Using Sequential Sparse Recovery
Improving the interpretability of deep neural networks with stimulated learning
HILK++: an interpretability-guided fuzzy modeling methodology for learning readable and comprehensible fuzzy rule-based classifiers
Visualizing the hidden activity of artificial neural networks
ActiVis: Visual Exploration of Industry-Scale Deep Neural Network Models
DeepEyes: Progressive Visual Analytics for Designing Deep Neural Networks
Discovering internal representations from object-cnns using population encoding
Axiomatic Attribution for Deep Networks
Detecting statistical interactions from neural network weights
Decoding the Deep: Exploring class hierarchies of deep representations using multiresolution matrix factorization
Explaining the Unexplained: A CLass-Enhanced Attentive Response (CLEAR) Approach to Understanding Deep Neural Networks
Explaining nonlinear classification decisions with deep taylor decomposition
Every Filter Extracts A Specific Texture In Convolutional Neural Networks
ShapeShop: Towards Understanding Deep Learning Representations via Interactive Experimentation
Generating interpretable images with controllable structure
Network Dissection: Quantifying Interpretability of Deep Visual Representations