Interpreting the Predictions of Complex ML Models by Layer-wise Relevance Propagation
By Wojciech Samek, Grgoire Montavon, Alexander Binder, Sebastian Lapuschkin, K. Müller. 2016
57 citations recorded by the source corpus.
Complex nonlinear models such as deep neural network (DNNs) have become an important tool for image classification, speech recognition, natural language processing, and many other fields of application. These models however lack transparency due to their complex nonlinear structure and to the complex data distributions to which they typically apply. As a result, it is difficult to fully characterize what makes these models reach a particular decision for a given input. This lack of transparency can be a drawback, especially in the context of sensitive applications such as medical analysis or security. In this short paper, we summarize a recent technique introduced by Bach et al. [1] that explains predictions by decomposing the classification decision of DNN models in terms of input variables.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Anand Raghunathan
- Kaushik Roy
- Sourjya Roy
- Sai Aparna Aketi
- Gobinda Saha
- Pietro Zanuttigh
- Umberto Michieli
Citations in the corpus
- A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI (2019, neutral): “[12] [35] [36] [37] [38] [39] [40] use Layer-wise Relevance Propagation (LRP) to construct heat-maps/relevance-maps for interpretability.”
- Gradual Channel Pruning While Training Using Feature Relevance Scores for Convolutional Neural Networks (2020, methodology): “The relevance scores of output nodes are then back propagated based on αβ-decomposition rule [31] with α = 2 and β = 1.”
- Explanations in Autonomous Driving: A Survey (2021, unknown): “[69], Layer-wise Relevance Propagation (LRP) [84], [85], DeepLift [86], [87], and Guided-Backpropagation [88].”