DATA ANALYTICS AND ARTIFICIAL INTELLIGENCE - 2

This course covers part of the "quantitative" aspects of the master program, and aims to introduce the student to the founding ideas of deep learning ad to offer a broad understanding of both its methodological and practical aspects.

More specifically, the goal of this course is to teach students the basic theoretical framework and to let them perform a practical, hands-on exploration of deep learning. Mathematical notation is paired with quantitative concepts via code snippets in order to help building practical intuition about the core ideas of machine learning and deep learning.

After the course, the students should have gained a solid understand of what deep learning is, when it’s applicable, and what its limitations are. They will be familiar with the standard
workflow for approaching and solving machine-learning problems, and they will know how to address commonly encountered issues.

Practice sessions will introduce to the use of Keras to tackle real-world problems ranging from computer vision to natural-language processing, image classification, timeseries forecasting, sentiment analysis, image and text generation, and more.

Students should have achieved the learning outcomes of courses "Computer Programming And Data Management" and they should have attended proficiently the first module of this course (even without passing the corresponding exams).

* Supervised and Deep Learning overview and models
* Training and assessing models
* Neural Networks from scratch
* Deep Learning from scratch
* Convolutional Neural Networks
* Recurrent Neural Networks
* Introduction to Keras
* Applications with Computer Vision and Text Processing

Deep Learning from Scratch: Building with Python from First Principles 1st Edition, Seth Weidman, O'Reilly

Learning outcomes are verified by a project.

The project requires to design and perform a deep learning task by both conceive the general framework and by gathering and preparing data to train the system. The task should be selected with the aim of addressing a real-world problem. Results must be demonstrated with both a written report and a live presentation.

scritto e orale

The exam will include a written test, which is mandatory, and a group project, which is optional and can allow to obtain up to three extra points.

Written exam (mandatory)

To be undertaken in presence at the end of the course (i.e. from the end of may).

This written exam will include three theoretical questions, about topics covered during the whole course, and a small exercise, asking to design from scratch a solution to a practical problem.

Each theoretical question will grant up to 4 points and the exercise will grant up to 20 points.

The students must submit the answers to the theoretical questions within 30 minutes form the start of the exam and the whole solution must be submitted within 90 minutes overall (i.e. if the student submits the theoretical questions in advance, he/she will have more time available for the exercise).

Group project (optional)

Students are encouraged to build groups from 1 to (max) 4 people. Each group must choose a groupname and send me a mail with (exactly) "[EM1405 Group Composition]" in the subject and the name and members of the groups in the text.

Each group will choose a data-driven problem of interest, will find the relevant data and will devise and implement all the tool needed to solve it. During the course several suggestions will be supplied, however the students are more that welcome to bring their very own ideas. Additionally, a few selected companies (especially during the second part of the course) will be invited to talk about relevant use cases and introduce challenges that can be addressed as a group project.

The group can submit the project any moment from the end of the course. There is not any time constraint for the presentation, however the final mark will be assigned when both the written exam and the project have been completed.

The submission must include:

A slide-based presentation of the activity (goals, methods, data used, approch, results)
Links to supporting materials (notebooks, data, etc.)

Should a student of the group decide to opt-out from the project, he/she can do anytime and get the grade of the written exam registered.

The group project will grant a maximum of 3 extra points to be added to the grading of the written exam.

Theoretical and practical lectures.
Exercise lectures.