ML Anthology

Visualizing Neural Networks with the Grand Tour

Mingwei Li, Zhenge Zhao, Carlos Scheidegger
Distill 2020
doi:10.23915/distill.00025 /distill/2020/li2020distill-visualizing/

Abstract

Distill articles are interactive publications and do not include traditional abstracts. This summary was written for the ML Anthology. Demonstrates the Grand Tour, a linear dimensionality reduction technique for visualizing neural network behavior that preserves interpretable structure better than non-linear alternatives like t-SNE. Applies it to observe training dynamics, trace data flow through layers, and analyze adversarial examples.

Distill Code Semantic Scholar

Cite

Text

Li et al. "Visualizing Neural Networks with the Grand Tour." Distill, 2020. doi:10.23915/distill.00025

Markdown

[Li et al. "Visualizing Neural Networks with the Grand Tour." Distill, 2020.](https://mlanthology.org/distill/2020/li2020distill-visualizing/) doi:10.23915/distill.00025

BibTeX

@article{li2020distill-visualizing,
  title     = {{Visualizing Neural Networks with the Grand Tour}},
  author    = {Li, Mingwei and Zhao, Zhenge and Scheidegger, Carlos},
  journal   = {Distill},
  year      = {2020},
  doi       = {10.23915/distill.00025},
  url       = {https://mlanthology.org/distill/2020/li2020distill-visualizing/}
}