• Publications
  • Influence
On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation
TLDR
We introduce a methodology that allows to visualize the contributions of single pixels to predictions for kernel-based classifiers over Bag of Words features and for multilayered neural networks. Expand
  • 1,166
  • 199
  • PDF
Explaining nonlinear classification decisions with deep Taylor decomposition
TLDR
We introduce a novel methodology for interpreting generic multilayer neural networks by decomposing the network classification decision into contributions of its input elements. Expand
  • 500
  • 58
  • PDF
Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment
TLDR
We present a deep neural network-based approach to image quality assessment (IQA). Expand
  • 279
  • 50
  • PDF
Methods for interpreting and understanding deep neural networks
TLDR
This paper provides an entry point to the problem of interpreting a deep neural network model and explaining its predictions. Expand
  • 747
  • 37
  • PDF
Evaluating the Visualization of What a Deep Neural Network Has Learned
TLDR
We present a general methodology based on region perturbation for evaluating ordered collections of pixels such as heatmaps. Expand
  • 404
  • 33
  • PDF
Explaining Recurrent Neural Network Predictions in Sentiment Analysis
TLDR
We propose a specific propagation rule applicable to multiplicative connections as they arise in recurrent network architectures such as LSTMs and GRUs. Expand
  • 149
  • 19
  • PDF
Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models
TLDR
This paper summarizes recent developments in this field and makes a plea for more interpretability in artificial intelligence. Expand
  • 401
  • 14
  • PDF
Divergence-Based Framework for Common Spatial Patterns Algorithms
TLDR
We propose a common divergence-based framework for spatial filter computation and introduce a general framework for this task based on divergence maximization. Expand
  • 132
  • 13
  • PDF
Stationary common spatial patterns for brain-computer interfacing.
TLDR
In this paper, we propose a method which regularizes CSP towards stationary subspaces (sCSP) and show that this increases classification accuracy, especially for subjects who are hardly able to control a BCI. Expand
  • 193
  • 12
  • PDF
Unmasking Clever Hans predictors and assessing what machines really learn
TLDR
We propose a semi-automated Spectral Relevance Analysis that provides a practically effective way of characterizing and validating the behavior of nonlinear learning machines. Expand
  • 179
  • 12
  • PDF