Zur Modulseite PDF generieren

#51049 / #1

SoSe 2023 - SoSe 2023

English

Applied Machine Learning in Engineering
Angewandtes Maschinelles Lernen im Ingenieurwesen

6

Stender, Merten

Benotet

Portfolioprüfung

English

Zugehörigkeit


Fakultät V

Institut für Maschinenkonstruktion und Systemtechnik (IMS)

35352200 FG Cyber-Physical-Systems in Mechanical Engineering

Maschinenbau

Kontakt


H 66

Stender, Merten

merten.stender@tu-berlin.de

Lernergebnisse

All engineering disciplines today employ machine learning for monitoring systems and fault detection, for data-based decision support as well as for leveraging new potentials in the environment of big data. This module teaches the fundamentals of standard machine learning techniques as well as their implementation using standard libraries in the Python programming language based on real-world engineering examples. Focus is put on the complete data science process from data exploration over modeling to inference and production. After successfully passing the module, students will have the following Knowledge: - Understanding of basic concepts of (un-) supervised machine learning and their structure and functionality - Comprehension of structure and conception of artificial neural networks - Familiarity with the basics of error backpropagation and optimization Skills: - Statistical characterization and evaluation of large and high-dimensional tabular data sets - Detection of outliers and anomalies - Appropriate visual representation of large and high-dimensional data sets - Programming basic operations and implementing regression and clustering models - Usage of well-known program libraries via the Python programming language Competencies: - Exploratory analysis of large multivariate tabular data sets. - Selection of machine learning techniques appropriate to the prediction purpose and complexity of the prediction task - Evaluation of machine learning methods with respect to goodness of fit and generalization behavior. - Risk and technological impact assessment The course teaches: 60% knowledge & understanding, 20% analysis & methodology, 20% programming.

Lehrinhalte

- Introduction to data-based methods and their applications in engineering. - Exploratory data analysis - Supervised regression models - Unsupervised clustering methods - Decision trees - Multilayer perceptrons and artificial neural networks - Gradient descent methods, error backpropagation, and training processes - Evaluation and assessment of machine learning methods - Practical examples from engineering disciplines - Programming tasks and implementation in the Python programming language - Risk and technological impact assessment

Modulbestandteile

Compulsory area

Die folgenden Veranstaltungen sind für das Modul obligatorisch:

LehrveranstaltungenArtNummerTurnusSpracheSWS ISIS VVZ
Applied Machine Learning in EngineeringVLSoSeen2
Applied Machine Learning in EngineeringUESoSeen2

Arbeitsaufwand und Leistungspunkte

Applied Machine Learning in Engineering (VL):

AufwandbeschreibungMultiplikatorStundenGesamt
Attendance15.02.0h30.0h
Pre/post processing15.04.0h60.0h
90.0h(~3 LP)

Applied Machine Learning in Engineering (UE):

AufwandbeschreibungMultiplikatorStundenGesamt
Attendance15.02.0h30.0h
Pre/post processing15.02.0h30.0h
Homework assignment2.015.0h30.0h
90.0h(~3 LP)
Der Aufwand des Moduls summiert sich zu 180.0 Stunden. Damit umfasst das Modul 6 Leistungspunkte.

Beschreibung der Lehr- und Lernformen

- Lecture: Class to convey the course content and contexts as frontal teaching with many examples from practice and interactive questions. - Exercise: practical and guided implementation of programming tasks in the programming language Python as well as exercises in small groups to deepen and apply the lecture material.

Voraussetzungen für die Teilnahme / Prüfung

Wünschenswerte Voraussetzungen für die Teilnahme an den Lehrveranstaltungen:

- Basic statistics - Advanced analysis (partial differentiation, gradient calculation) and linear algebra (matrix and tensor multiplication, projection methods, matrix decomposition). - Basic concepts and methods of sequential/object-oriented programming, optimally in Python or Matlab

Verpflichtende Voraussetzungen für die Modulprüfungsanmeldung:

Dieses Modul hat keine Prüfungsvoraussetzungen.

Abschluss des Moduls

Benotung

Benotet

Prüfungsform

Portfolio examination

Art der Portfolioprüfung

100 Punkte insgesamt

Sprache(n)

English

Prüfungselemente

NamePunkteKategorieDauer/Umfang
Homework assignment 120schriftlich3 pages
Homework assignment 220schriftlich3 pages
Oral exam60mündlich30min

Notenschlüssel

Notenschlüssel »Notenschlüssel 7: Fak V«

Gesamtpunktzahl1.01.31.72.02.32.73.03.33.74.0
100.0pt89.0pt85.0pt80.0pt76.0pt72.0pt67.0pt63.0pt59.0pt54.0pt50.0pt

Prüfungsbeschreibung (Abschluss des Moduls)

The portfolio examination is composed of an oral exam (at the end of the semester) and two homework assignments (to be handed in during the semester). Homework assignment 1: Exploratory data analysis of a dataset provided to the students. A written report of 3 pages (typesetting template will be provided) including a short textual discussion and graphical visualizations is to be handed in within 3 weeks after the announcement. Homework assignment 2: Implementation of a basic modeling technique in Python programming language and evaluation on a given data set. A written report of 3 pages (typesetting template will be provided) including a short textual discussion, code snippet, and graphical visualization of the result is to be handed in within 3 weeks after the announcement. Oral exam: Examination of the lecture contents, general concepts, categorization of machine learning concepts, and discussion of elements of the machine learning life cycle process.

Dauer des Moduls

Für Belegung und Abschluss des Moduls ist folgende Semesteranzahl veranschlagt:
1 Semester.

Dieses Modul kann in folgenden Semestern begonnen werden:
Sommersemester.

Maximale teilnehmende Personen

Die maximale Teilnehmerzahl beträgt 30.

Anmeldeformalitäten

Registration for the examination according to AllgStuPO in QISPOS; access to teaching material and registration for the course via the e-learning platform ISIS.

Literaturhinweise, Skripte

Skript in Papierform

Verfügbarkeit:  nicht verfügbar

 

Skript in elektronischer Form

Verfügbarkeit:  verfügbar

 

Literatur

Empfohlene Literatur
Keine empfohlene Literatur angegeben

Zugeordnete Studiengänge


Diese Modulversion wird in folgenden Studiengängen verwendet:

Studiengang / StuPOStuPOsVerwendungenErste VerwendungLetzte Verwendung
Dieses Modul findet in keinem Studiengang Verwendung.

Sonstiges

Students are asked to use their own computers for the programming tasks. The oral examination can also be held in German if preferred by the student.