Learning from Unstructured Data


At the end of this unit, students will:

-Basic principles of deep learning.
-Unsupervised extraction of features and learned representations for use in regression or classification in multi-layer models.
-Optimization and regularization methods applicable to models with a large number of parameters
-The different models presented: fully connected feed-forward neural networks, convolution networks, and recurrent networks.
-Problems and techniques for processing unstructured data.

Be able to:
Select and implement models to solve some typical problems and optimize training to obtain a reasonable solution

Problems solvable with deep learning: object recognition in images, voice recognition, natural language processing, and others.

General characterization





Responsible teacher

Claudia Alexandra Magalhães Soares, João Alexandre Carvalho Pinheiro Leite


Weekly - 4

Total - 48

Teaching language





-Ian Goodfellow, Yoshua Bengio, Aaron Courville, Deep Learning, MIT Press

-Yann LeCun , Yoshua Bengio and Geoffrey Hinton, Deep learning, Nature, Vol. 521, pp 436-444, 2015

Teaching method

Lectures will cover the fundamental topics of the subject matter, which the students should complement with the given bibliography. All lecture materials will be supplied for further study. Lectures will include some time for questions and discussion of the subject matter.

Tutorial classes will be dedicated to exercises and guidance in the practical assignments, focusing on selected topics.

Class schedules and materials will be supplied online, as well as additional information regarding the course.

Evaluation method

The final grade is the weighted average of:

30%: theory assessment.

20%: First project.

50%: Theory and practice assessment.

Subject matter

1-Introduction: data types and problems.
2-Deep Feedforward networks
3-Optimization of Feedforward networks.
4-Regularization of Feedforward networks.
5-Convolution networks.
6-Recurrent networks and sequential problems.
7-Introduction to generative models.
8-Transfer learning
9- Additional problems and examples.