Zur Modulseite PDF generieren

#41135 / #4

WiSe 2023/24 - WiSe 2023/24

English

ROC Foundations for Graduate Research in Data Management and Machine Learning Systems

9

Markl, Volker

Benotet

Portfolioprüfung

English

Zugehörigkeit


Fakultät IV

Institut für Softwaretechnik und Theoretische Informatik

34351500 FG Datenbanksysteme und Informationsmanagement (DIMA)

Keine Angabe

Kontakt


EN 7

Markl, Volker

sekr@dima.tu-berlin.de

Lernergebnisse

Big Data and Machine Learning are key drivers underlying the current wave of innovation in artificial intelligence and data science. Indeed, these drivers have had a profound impact on both the economy and the sciences. This course targets research-oriented students who aim to pursue a PhD in Big Data Management -or- Data Science and Engineering Systems/Technologies. Upon completion of this course, students will have learned about contemporary research methodology, including scientific reading, writing, presenting, prototyping and experimental design, gained both theoretical and practical skills in data management and big data technologies, and be attuned to today’s major research challenges in scalable data management and processing. The course is designed to principally impart technical skills (20%), method skills (40%), systems skills (20%), and social skills (20%).

Lehrinhalte

The central focus of this module is on contemporary research methodology (CRM), data management technologies, and current research challenges. After an initial presentation on CRM, including scientific reading, writing, presenting, prototyping and experimental design, in subsequent lectures, students will read about foundational data management methods/ technologies and offer a presentation, which will then be followed by an instructor led presentation addressing related advanced topics. Topics of discussion, include data storage and indexing, specification and compilation of data analysis programs, query optimization and self-tuning, adaptive methods, processing data science pipelines as well as responsible data management. In an accompanying lab component, students will prototype and evaluate discussed methods, technologies, and settings in a methodical and scientific way, and produce a scientific report on their findings.

Modulbestandteile

Compulsory area

Die folgenden Veranstaltungen sind für das Modul obligatorisch:

LehrveranstaltungenArtNummerTurnusSpracheSWS ISIS VVZ
ROC-PRO Project on the Foundations for Graduate Research in Data Management and Machine Learning SystemsPJWiSeen4
ROC-SEM Seminar on the Foundations for Graduate Research in Data Management and Machine Learning SystemsSEMWiSeen2

Arbeitsaufwand und Leistungspunkte

ROC-PRO Project on the Foundations for Graduate Research in Data Management and Machine Learning Systems (PJ):

AufwandbeschreibungMultiplikatorStundenGesamt
Lab course (Programming)15.02.0h30.0h
Lab Course (System Setup)15.02.0h30.0h
Lab Course (Experimental Setup)15.02.0h30.0h
Report15.02.0h30.0h
Lab course (Performance Evaluation)15.04.0h60.0h
180.0h(~6 LP)

ROC-SEM Seminar on the Foundations for Graduate Research in Data Management and Machine Learning Systems (SEM):

AufwandbeschreibungMultiplikatorStundenGesamt
Plenary Sessions15.04.0h60.0h
Preparation and Presentation (including reading and literature research)15.02.0h30.0h
90.0h(~3 LP)
Der Aufwand des Moduls summiert sich zu 270.0 Stunden. Damit umfasst das Modul 9 Leistungspunkte.

Beschreibung der Lehr- und Lernformen

This module (comprised of ROC-PRO and ROC-SEM) encompasses: (a) lectures on key concepts, (b) discussions, (c) student lead presentations (including literature search), and (d) a systems research project including system setup, prototyping, experimental design, performance evaluation, and (e) creating a presentation and report on the findings. Active participation and contributions to all parts of ROC are essential.

Voraussetzungen für die Teilnahme / Prüfung

Wünschenswerte Voraussetzungen für die Teilnahme an den Lehrveranstaltungen:

Desired prerequisite knowledge and skills are as follows: (a) computer science topics addressed in TU Berlin modules in the Bachelor’s curriculum, particularly, ISDA (Information Systems and Data Analysis) and DBPRA (Practical Database Systems Lab) or their equivalents, (b) good programming skills in C, Java, and SQL. (c) an undergraduate course in linear algebra, probability, and statistics. (d) knowledge of a master's level coursework in database technology (DBT) and advanced information management (e.g., MDS, DMH). (e) strong English language skills.

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
Effective and Efficient Interaction with the Mentor (PRO)10flexibel30 to 60 minutes per need
Evaluation Report (PRO)50schriftlich8 pages, conference style
Interaction with the Mentor (SEM)10flexibel30 to 60 minutes per need
Performance Evaluation Presentation (PRO)40mündlich20 min + 10 min discussion
Quiz on Database Technology and Research Methodology (SEM)50schriftlich60 minutes
Technology Presentation (SEM)40mündlich20 min + 10 min discussion (20 Slides)

Notenschlüssel

Notenschlüssel »Notenschlüssel 2: Fak IV (2)«

Gesamtpunktzahl1.01.31.72.02.32.73.03.33.74.0
100.0pt95.0pt90.0pt85.0pt80.0pt75.0pt70.0pt65.0pt60.0pt55.0pt50.0pt

Prüfungsbeschreibung (Abschluss des Moduls)

For ROC-PRO the exam is worth 100 points and determined as follows: evaluation report (50 points), interaction with the mentor (10 points), and a performance evaluation presentation (40 points). For ROC-SEM the exam is worth 100 points and determined as follows: technology presentation (40 points), quiz on database technology and research methodology (50 points), and the interaction with the mentor (10 points). For both ROC-PRO and ROC-SEM the final grade will be computed according to the Grading Table 2 of Faculty IV, according to German law, § 68 (2) AllgStuPO TU Berlin.

Dauer des Moduls

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

Dieses Modul kann in folgenden Semestern begonnen werden:
Wintersemester.

Maximale teilnehmende Personen

Die maximale Teilnehmerzahl beträgt 8.

Anmeldeformalitäten

Students are required to register for the course in the official TUB examination system within six weeks after commencement of the first lecture or when the first graded assignment is due, whichever happens to be first.

Literaturhinweise, Skripte

Skript in Papierform

Verfügbarkeit:  nicht verfügbar

 

Skript in elektronischer Form

Verfügbarkeit:  nicht verfügbar

 

Literatur

Empfohlene Literatur
Readings in Database Systems, 5th Edition, Peter Bailis, Joseph M. Hellerstein, Michael Stonebraker, editors, http://www.redbook.io/
Various Research Papers, made available during the first lecture
Hadoop: The Definitive Guide (4th Edition), Tom White, O’Reilly Media, 2015.
Raj Jain: The Art of Computer Systems Performance Analysis: Techniques for Experimental Design, Measurement, Simulation, and Modeling (Wiley Professional Computing), 1991
Supplementary reading material may be assigned to complement course lectures.

Zugeordnete Studiengänge


Diese Modulversion wird in folgenden Studiengängen verwendet:

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

Sonstiges

This course targets research-oriented Bachelor’s and Master’s students interested in focusing on Database Systems and Information Management in Computer Science (Major: System Engineering), Computer Engineering (Major: Information Systems and Software Engineering), and Industrial Engineering, as well as students pursuing the Data Science and Engineering Master’s Track.