Inhalt

Abstract:

This course focuses on various aspects of Deep Learning. Theoretical foundations and general concepts are introduced in the first part, while the second part focuses on specific networks used in image analysis as well as time-series analysis, two common tasks in engineering applications.
The list of topics covered includes:
- Network optimization
- Regularization
- Convolutional neural networks
- Reccurent neural networks
In the integrated lab sessions, the students will tackle an image classification problem as well as a time-series regression problem using industrial datasets.

Lern- und Qualifikationsziele

After completion of this course, students should be capable of chosing and implementing a suited deep learning algorithm for a given problem based on the type of data and the general learning task.

Gliederung:

At the beginning, a brief overview of the essential concepts of machine learning will be given as a refresher, but no detailed explanation of the content will be given as this is already covered in the ML4Engineers I course. A basic understanding of machine learning and programming, particularly in Python, is recommended to participate in this advanced course. The previous ML4Engineers I provides the ideal foundation for this, especially because of the aligned curriculum.

Following is an introduction to the theory of Deep Learning and the different types such as Convolutional (CNN) and Long-Short-Term Memory (LSTM). Students will learn how to solve complex problems using such methods. Further, they will discuss how these methods are applied to different types of data, e.g., text, image, audio files.

The second block presents different applications of Deep Learning in practice. First, students learn which tools and libraries are available for such methods and how to use them. Tensorflow and Keras are the focus here as established tools for Deep Learning applications. After the students have been taught how to use these methods by means of examples, they are practiced through various programming exercises. Here, attention is paid to the heterogeneity of the tasks (anomaly detection, time series prediction, etc.) as well as the data basis (image, sound, text), which should enable the students to apply the presented methods in different scenarios.

Detaillierter Inhalt:

Lectures:

• Introduction to Machine Learning
• Deep Learning Basics - Part 1
• Deep Learning Basics - Part 2
• Optimization
• Regularization
• Convolutional Neural Networks - Part 1
• Convolutional Neural Networks - Part 2
• Recurrent Neural Networks - Part 1
• Recurrent Neural Networks - Part 2

Labs:

• Optimization
• Image Classification and Transfer Learning
• Timeseries Analysis

Schwierigkeitsgrad:

Erfahrene

Lehr-/Lernform:

Kurs

Interaktionsformen mit dem System/Betreuer:

E-Mail

Interaktionsformen mit Mitlernenden:

Forum

Kursdemo:

zur Kursdemo

Schlagworte:

Engineering, Informatik, Deep Learning, Datenanalyse, Künstliche Intelligenz, Maschinelles Lernen, Klassifikation, Regression

Nutzung

Zielgruppe:

Ing.-Wissenschaften für Uni-Studierende, Ing.-Wissenschaften für FH-Studierende

Nutzbar im Studiengang:

-

Geeignet für Berufsfeld:

-

Formale Zugangsvoraussetzungen:

-

Erforderliche Vorkenntnisse:

No prior knowledge is mandatory, but the understanding of basic machine learning concepts and coding skills is strongly advised, e.g. by participcating in the course "Machine Learning for Engineers: Introduction to Methods and Tools"

Erforderliche Vorkenntnisse bzgl. Handhabung der Lernplattform:

-

Verantwortlich

Trägerhochschule:

Uni Erlangen-Nürnberg (FAU)

Anbieter:

Prof. Dr. Björn Eskofier

Autoren:

Jörg Franke, Nico Hanenkamp, Björn Eskofier

Betreuer:

Thomas Altstidl

Prüfung

Online course exam for students

Art der Prüfung:

schriftlicher Leistungsnachweis (Klausur)

Prüfer:

Prof. Dr. Björn Eskofier

Anmeldeverfahren:

–

Prüfungsanmeldefrist:

–

Prüfungsabmeldefrist:

–

Kapazität:

–

Prüfungsdatum:

–

Prüfungszeitraum:

–

Prüfungsdauer:

60 Minuten

Prüfungsort:

–

Zustündiges Prüfungsamt:

Anerkennung: Heimathochschule der Studierenden

Zugelassene Hilfsmittel:

–

Formale Voraussetzungen für die Prüfungsteilnahme:

–

Inhaltliche Voraussetzungen für die Prüfungsteilnahme:

–

Zertifikat:

Ja (benoteter Schein)

Anerkennung an folgenden Hochschulen:

Uni Erlangen-Nürnberg (FAU), FH Aschaffenburg, FH Rosenheim, FH Amberg-Weiden

Sonstige Anerkennung:

noch nicht bekannt

Online-Prüfungsan-/-abmeldung:

Nein

Bemerkung:

The examination is online and the date is announced in the course.

Erforderliche Technik

Verwendete Lernplattform:

StudOn (ILIAS)

Nutzungsbedingungen

Gebühren:

Nein

Nutzungsentgelte:

für andere Personen als (reguläre) Studenten der vhb Trägerhochschulen nach Maßgabe der Benutzungs- und Entgeltordnung der vhb

Copyright:

-

Hinweise zur Nutzung:

-