Diagrammatic Summaries for Neural Architectures

Abstract

This paper advocates for diagrammatic summary publications for machine learning system architecture papers. We review existing diagram-centric scholarly practices, and summarise relevant studies on neural network system architecture diagrams. We subsequently propose three opportunities: Diagram guidelines, diagrammatic system summary publications, and the community creation of a formal diagram standards, which could be integrated with existing LaTeX + PDF publication processes.

Cite

Text

Marshall et al. "Diagrammatic Summaries for Neural Architectures." ICLR 2021 Workshops: Rethinking_ML_Papers, 2021.

Markdown

[Marshall et al. "Diagrammatic Summaries for Neural Architectures." ICLR 2021 Workshops: Rethinking_ML_Papers, 2021.](https://mlanthology.org/iclrw/2021/marshall2021iclrw-diagrammatic/)

BibTeX

@inproceedings{marshall2021iclrw-diagrammatic,
  title     = {{Diagrammatic Summaries for Neural Architectures}},
  author    = {Marshall, Guy Clarke and Jay, Caroline and Freitas, André},
  booktitle = {ICLR 2021 Workshops: Rethinking_ML_Papers},
  year      = {2021},
  url       = {https://mlanthology.org/iclrw/2021/marshall2021iclrw-diagrammatic/}
}