Turing Completeness of Neural Network Architectures: a Review

This repository has been created with the sole objective of sharing a comprehensive literature review about the expressive capabilities of different Neural Network architectures from a theoretical perspective.

Table of Contents

• Introduction
• Repository Structure
• Getting Started
• Article
• Presentation
• Citation
• License

Introduction

This repository provides resources related to the study of Turing Completeness in Neural Networks. The materials include a comprehensive literature review about this topic, along with an accompanying presentation suitable for an academic seminar.

Repository Structure

The repository is organized as follows:

. article article.pdf src main.tex references.bib presentation slides.pdf src cgru.jpg gru.jpg slides.md theme.css transformers.png README.md LICENSE CITATION.cff

• article/: Contains the article PDF and its source files.
• presentation/: Contains the presentation PDF and its source files.
• README.md: This README file.
• LICENSE: The license under which this project is distributed.
• CITATION.cff: Citation information for this repository.

Getting Started

To explore the contents of this repository, follow the steps below:

1. Clone the Repository:

bash git clone https://github.com/dambrosidenis/Turing_Completeness_of_Neural_Network_Architectures.git cd Turing_Completeness_of_Neural_Network_Architectures

1. View the Article:

Open the article/article.pdf file to read the full article.

1. View the Presentation:

Open the presentation/slides.pdf file to review the presentation.

Article

The detailed article explaining the Turing Completeness of neural networks can be found in the article/ directory:

• Article PDF

The source files for the article, including LaTeX and bibliography, are located in article/src/. The article can be compiled through pdflatex and bibtex.

Presentation

The presentation summarizing the key findings can be found in the presentation/ directory:

• Slides PDF

The source files for the presentation, including images and styles, are located in presentation/src/. The presentation is written in markdown format using the Marp extension.

Citation

If you use this work in your research, please cite it as follows:

@software{D_Ambrosi_Turing_Completeness_of_2023, author = {D'Ambrosi, Denis}, month = feb, title = {{Turing Completeness of Neural Networks Architectures: a Review}}, url = {https://github.com/dambrosidenis/Turing_Completeness_of_Neural_Network_Architectures/}, version = {1.0.0}, year = {2023} }

License

This project is licensed under the MIT License. See the LICENSE file for more details.