Deep Learning - Interaktive Studienpläne der TU Ilmenau

Das Modul Deep Learning mit der Prüfungsnummer 220488 schließt mit folgenden Leistungen ab:

• alternative semesterbegleitende Prüfungsleistung mit einer Wichtung von 50% (Prüfungsnummer: 2200822)
• alternative semesterbegleitende Prüfungsleistung mit einer Wichtung von 50% (Prüfungsnummer: 2200823)

Details zum Abschluss Teilleistung 1:

• multiple coding assignments evaluating methodological and practical competence in the taught concepts - to be individually solved at home with due date and submission via Moodle
• result determined as average across the evaluated solutions to the assignments
• students must register via thoska for this exam, typically within the 3rd and 4th week of the semester

 Details zum Abschluss Teilleistung 2:

• one or multiple written tests consisting of multiple-choice and free-form questions evaluating the professional competence in the course's topics
• preferably conducted digitally via Moodle and on the student's device
• final results may be scaled or individual questions may be excluded depending on best performing percentile of students
• students must register via thoska for this exam, typically within the 3rd and 4th week of the semester

Dieses Modul enthält mindestens eine alternative semesterbegleitende Abschlussleistung. Bitte beachten Sie, dass diese in der Regel schon zu Beginn des Semesters, in dem diese angeboten wird, angemeldet werden muss. Über die Details und Zeiträume dazu werden Sie vom Lehrenden und/oder dem Prüfungsamt informiert. Fragen Sie gegebenenfalls unbedingt beim Lehrenden nach.

This module contains at least one alternative exam part. Please note that this must usually be registered at the beginning of the semester in which it is offered. The lecturer and/or the examination office will inform you about the details and time periods. If necessary, be sure to ask the lecturer.

• programming knowledge (Python)
• foundations of mathematics (linear algebra and calculus)
• foundations of machine learning

Professional competence gained through lectures and examined through written exam:

• Students have knowledge about theoretical foundations of deep neural networks.
• Students have knowledge about CNN architectes and their applications.
• Students have knowledge about architectures for sequence modeling and their applications.

Methodological competence gained through seminars and examined through aPl (assignments):

• Students gained the ability to implement and apply a variety of deep learning algorithms.
• Students gained the ability to evaluate and troubleshoot deep learning models.
• Students gained the ability to use computational resources for training and application of deep learning models.

Social competence gained through lectures and seminars:

• Students gained insights in ethical aspects of machine learning (e.g., bias, autonomous driving) through discussions in lectures and seminars.
• Students can discuss advantages and disadvantages of different deep learning approaches among each other and with their lecturers and gained professionality in mastering discussions beyond their mother tongue.
• Students learn to discuss and solve a scientific problem in a team of peers

Deep learning has revolutionized a variety of application like speech recognition, image classification, language translation, as well as text and image generation. Today, deep learning techniques are applied for solving an ever-increasing variety of problems. This course will give you detailed insight into deep learning, introducing you to the fundamentals as well as tools and methods in this rapidly emerging field.

Deep learning thereby refers to a subset of machine learning algorithms that analyze data in succeeding stages, each operating on a different representation of the analyzed data. Specific to deep learning is the ability to automatically learn these representations rather than relying on domain expert for defining them manually.

The course will teach you the theoretical foundations of deep neural networks, which will provide you with the understanding necessary for adapting and successfully applying deep learning on your own to implement, parametrize and apply a variety of neural network architectures for modelling different data modalities and solving a variety of problems.

• Presentations
• Homework task sets as PDF
• Assignments including code stubs
• Jupyter notebooks
• All material will be shared via Moodle, accessible [HERE]

Technical Requirements

• available access to moodle.tu-ilmenau.de
• available access to colab.google.com

• Deep Learning: Ian Goodfellow, Yoshua Bengio, and Aaron Courville, MIT Press (2016)
  

• Pattern Recognition and Machine Learning: Christoper M. Bishop, Springer (2006)
  

• Hands-On Machine Learning with Scikit-Learn and TensorFlow: Aurélien Géron, O'Reilly Media (2017)