TABLE OF CONTENTS

Foundations of Artificial Intelligence:

Neural Networks and Transformers

Department of Mathematics and Statistics, Universidad del Norte, Barranquilla, Colombia

hllinas@uninorte.edu.co

Biographical sketch

05/02/26

Abstract

Other related documents can be found at Rpubs:: toc.

Preface

Contemporary artificial intelligence is grounded in a rigorous mathematical framework that enables language to be represented, transformed, and learned from. From the construction of vocabularies and the formalization of text as mathematical objects, to neural networks and Transformer architectures that dominate modern natural language processing, each development is supported by well-established principles from statistics, linear algebra, and geometry.

This document presents a structured overview of the mathematical foundations underlying language representation, with particular emphasis on how these ideas are integrated into modern artificial intelligence models. The table of contents that follows guides the reader from basic lexical concepts to vector-based representations that make large-scale machine learning possible.

Source: Vaswani et al. (2017)

1 Mathematics behind language representation

1.1 Lexical foundations and vocabulary construction in NLP: Right-click here.

1.2 Transforming text into data structures: Right-click here.

1.3 Vectorizing text and quantifying similarity: Right-click here.

References

 

 

If you found any ERRORS or have SUGGESTIONS, please report them to my email. Thanks.