MLLGJun 19, 2023

Interpreting Deep Neural Networks with the Package innsight

arXiv:2306.10822v29 citationsh-index: 18
Originality Synthesis-oriented
AI Analysis

This package addresses the need for interpretability tools in deep learning for R users, though it is incremental as it adapts existing methods to a new software environment.

The authors developed the R package innsight, a toolbox for interpreting deep neural network predictions using feature attribution methods, which works with models from various R packages and provides visualization tools for different data types.

The R package innsight offers a general toolbox for revealing variable-wise interpretations of deep neural networks' predictions with so-called feature attribution methods. Aside from the unified and user-friendly framework, the package stands out in three ways: It is generally the first R package implementing feature attribution methods for neural networks. Secondly, it operates independently of the deep learning library allowing the interpretation of models from any R package, including keras, torch, neuralnet, and even custom models. Despite its flexibility, innsight benefits internally from the torch package's fast and efficient array calculations, which builds on LibTorch $-$ PyTorch's C++ backend $-$ without a Python dependency. Finally, it offers a variety of visualization tools for tabular, signal, image data or a combination of these. Additionally, the plots can be rendered interactively using the plotly package.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes