AI_Workshop

Welcome to this workshop on artificial intelligence! In the following days, we will explore both basic and advanced techniques in machine learning, artificial neural networks, and deep neural networks. We will cover the theory behind these methods and provide hands-on practice using the Python programming language.

About the instructors

• Dr. Wilfrido Gomez-Flores obtained a BS in Electronics and Communications Engineering from the Technological University of Mexico in 2004. He then earned an M.Sc. (2006) and D.Sc. (2009) in Electrical Engineering from Cinvestav. Since 2010, he has been a researcher at Cinvestav Tamaulipas Campus. Over the course of his career, he has produced over 100 publications, graduated 14 master's students and four doctoral students, and is a Level 2 in the SNII. His research interests include digital image analysis, pattern recognition, and machine learning.

• Dr. Fernando Arce-Vega holds a Bachelor's degree in Electronic Engineering and a Master's degree in Bioelectronics from Cinvestav. He completed his PhD in Computer Science at CIC-IPN and holds more than 24 certifications in machine learning and deep learning, as well as several scientific publications. As a Level 1 member of the National System of Researchers (SNII), he focuses on developing AI algorithms for optimizing biosensors and generating automatic mathematical models.

Workshop topics

Introduction: Introduction to the Workshop

Part 0: Introduction to the Programming Environment

• Google Colab
• Tensors

Part 1: Machine Learning

• What is artificial intelligence?
• Introduction to machine learning
• Basic concepts and definitions
• Supervised learning models
  • Linear regression
  • Naive Bayes
  • KNN
  • Logistic regression
• Unsupervised learning models
  • PCA
  • K-means
• Evaluating classification models

Part 2: Neural Networks

• Biological inspiration
  • Perceptron
• Gradient descent
• Single-layer neural networks
  • MLP
• Multi-layer perceptron
  • MLP-SGD
• SGD and optimizers
  • MLP-complete
  • MLP-autoencoder
• Universal approximation theorem

Part 3: Deep Neural Networks

• What is deep learning?
• Convolutional neural networks
• LeNet-5 model
  • CNN-MNIST
• Pre-trained CNN models
  • Transfer learning
• Instance segmentation
  • YOLO
• GRAD-CAM methods
  • GRAD-CAM
  • CNN-autoencoder

Recommended bibliography

Contact information

• Dr. Wilfrido Gomez-Flores: wgomez@cinvestav.mx
• Dr. Fernando Arce-Vega: farce@cio.mx