Artboard 1 copy 7.png
Parametric and non-parametric algorithms.

Objectives

1. To familiarize the student with the ideas and some of the most effective methods and in Data Science using two fundamental examples: Deep learning and decision trees.

2. That the student is able to understand the theoretical background of the chosen algorithms in order to offer autonomy to study and take advantage of their qualities.

3. Invite the student to the ideas and formalism of more elaborate techniques such as: pruning, stochastic gradient descent or back-propagation.

Syllabus

Module 1: Neural Networks

  1. Linear perceptron

  2. Gradient method

  3. Stochastic Gradient Descent

  4. Perceptron with more than one layer

  5. Back-propagation

Module 2: Decision Trees

  1. Overfitting vs fitting

  2. Decision trees

  3. Entropy, Gini and information measures (Shannon)

  4. Random Forests

  5. Pruning

Contact

CDMX, México.

  • Facebook icono social