General Information
Objective: Study of the main deep learning methods and their applications.
Syllabus: Machine learning basics; Deep Feedforward Networks; Convolutional Neural Networks (CNNs), Recurrent Neural Networks; Generative Adversarial Networks (GANs); Object Detection; Segmentation; Natural Language Processing (NLP); Advanced models; Frameworks for deep learning and Practical Aspects; Applications of deep learning models for real-world problems.
Duration/credits: 45 hours/3 credits (12 weeks).
Time: Tuesdays (17:50 - 21:10) - online with synchronous online (SYN) and assynchronous (ASYN) classes.
Lecturers: Heitor Silvério Lopes and André Eugenio Lazzaretti.
Collaborators: Andrei Inácio, Clayton Hilgemberg da Costa, Anderson Brilhador, Lucas da Silva Nolasco.
Prerequisites: Linear algebra; Probability and statistics; Differential and integral calculus; Programming (Python).
Python Basics (Lecture 3 and 4 from this link; Lectures 1.2, 1.3, 1.4, 1.5, 2.1, 2.2, 2.3, 2.4, and 2.5 from this link).
Math for ML (link).
Grade: Assignments (30%) and Final Project (70%).
Grades
Assignment and Final Project (link).
Final Project Rules
Final Project - Rules:
It can be in pairs.
Something more than getting data and network already available is expected.
Important to detail and interpret the results.
Important: it could become a publication in a conference/journal.
Project Proposals - Guidelines:
What is the problem to be solved? Preferably a real problem that can be solved via deep learning.
Appropriate answer: perform classification of EEG signals to identify movement intentions.
Inappropriate answer: I have this dataset (own, Kaggle, etc.) and intend to use a CNN to check the classification result.
Dataset available? Few data may make it unfeasible.
Do you need HW (GPU)? The LABIC cluster may be made available.
What techniques do you intend to use and why?
If you have a comparison parameter (papers, kaggle, etc).
Final Project - Presentation:
Final Report Deadline: Mar 08 (2024).
Must contain: Introduction (detailing the addressed problem); Methodology; Results (with comparisons, if possible); Conclusions.
Suggested template: IEEE.
Presentation Deadline: 14-15 Mar (2024). 15 minutes for each work + 5 minutes for questions.
Bibliography and Support Materials
Books:
Goodfellow, I., Bengio, Y., Courville, A. Deep Learning. MIT Press, 2016.
Chollet, F., Deep Learning with Python. Manning, 2018
Other Courses:
Andrew Ng at Coursera.
Sebastian Raschka at UW Madison.
Fei Fei Li et al. at Stanford (most similar).
Week 7 - November 21
Lecturer:
André, Heitor e Andrei.
Content:
Assignment 05:
(ASYN) All the students must read the original transformer paper. For the discussion in the next class:
List the questions and points that you did not understand;
Bring the aspects that you considered most relevant to the model;
Point out which extra materials you have consulted to understand the concepts.
Week 10 - December 12
Lecturer:
André e Heitor.
Content:
(SYNC) Final Project Discussions.