Artboard 7 copy 20.jpg
Deep Learning in Images and NLP: its Mathematics and implementation in Python

This online course consists of three modules on the fundamentals and applications of neural networks to image processing and NLP.

General information

  • Neural networks are one of the most powerful tools of Artificial Intelligence. Their use has changed the world we live in and it is difficult to think of an industry that does not use them.

  • In this course we will teach students to understand and use neural networks in detail, we will focus on four fundamental parts:

  • Implementation in Python and good practices

  • Description of the different algorithms

  • The different and differences between architectures

  • The most relevant mathematical ideas

  • The material is not strictly sequenced and for example if you have some notions about neural networks it is possible to start in the second one.

  • A deep knowledge of Python is not necessary, however it is recommended to be familiar with basic programming concepts.

Details and contact

-The event will take place remotely.

-Each module includes some notes in the form of a log about what has been learned.


Alfonso Ruiz

Tel. +52 1 55 59957954

The course will be taught in proportional times by the following teachers.

  • Beatriz Londoño is the founder of "Science and Technology for all". PhD in Physics from the Université Paris-XI Orsay with research experience and co-author of scientific articles. Understanding a variety of statistical analysis, programming, machine and deep learning methods. Teaching experience. Strong interest in science education and in making analysis and critical thinking a tool that is close to everyone.

  • Eduardo H. Ramírez is a Doctor in Intelligent Systems from ITESM. He collaborated with Yahoo Research and Microsoft Research in Web Spam detection and Search Quality. He is the founder of the Data Science community in Monterrey, a member of the Caintra CVT council and, a global initiative aimed at empowering people to acquire AI and ML skills applied in social impact projects.

  • Alfonso Ruiz is a mathematician specialized in mathematical logic and its applications to other areas as well as a mathematics teacher. He studied his doctorate in mathematics at Oxford University, his master's degree at Université Paris-XI Orsay and a degree in mathematics at UNAM. He was professor of Mathematical Analysis at Corpus Christi College, Oxford. He is currently a principal at the Bourbaki School of Mathematics.



Module 1: Basics of Artificial Neural Networks

The first module is an introduction to neural networks that seeks to deepen a clear understanding of the benefits and how these techniques work. The most useful algorithms to train them will be presented and an explanation of the mathematics that explains them will be given.

  • Python reinforcement

  • Perceptron

  • Gradient method

  • Programming the gradient method

  • Stochastic gradient descent

  • Perceptron with more than one layer

  • Back-propagation

  • Back-propagation scheduling

  • Libraries and implementation tips

  • Case study on image classification

Module 2: NLP Architectures and Image Processing

In this module we will expose the main techniques and difficulties behind the use of neural networks to process images and text. We will teach how to use the main algorithms to build neural networks sensitive to the problems of the course. We will make special emphasis on the concept of convolution and mathematical development.

  • Historical development and general architectures of ANN

  • Main tasks of ANN architectures

  • State of the art architectures for image recognition (ImageNet, ResNet, GoogleNet)

  • Transfer Learning

  • Convolution of functions and distributions

  • Image classification using CNN

  • Detecting objects in scenes using CNN

  • Text translation: basic ideas

  • transformer

  • Module 3: Topics in Deep Learning

  • We will explain some useful ideas for image and text processing. For each of the concepts listed, we will begin with its formal description and continue with its study of the two problems that interest us.

  • Regularization in image classification and word processing.

  • Regularity in image classification and word processing.

  • Generative Models in image classification and text processing.





  • We offer 2 full scholarships per module one will be for a woman and the other will be for a man. It is necessary not to have a full-time job to apply for the scholarship. Please write to a reason letter explaining why you are interested in the course.


CDMX, México.

  • Facebook icono social