Moses - Machine Intelligence I

Display language

Diese Modulbeschreibung gilt nicht im aktuellen Semester.

Die Modulbeschreibung gilt nur für den Zeitraum SS 2018 - SS 2019. Die Version für das aktuelle Semester (SoSe 2024) finden Sie hier: Modulbeschreibung des SoSe 2024

Module / Version:
#40548 / #5

Validity:
SS 2018 - SS 2019

Default display language:
English

Module title:
Machine Intelligence I

Credits:
6

Responsible person:
Obermayer, Klaus

Grading:
benotet

Type of exam:
Schriftliche Prüfung

Zugehörigkeit

Faculty:
Fakultät IV

Institute:
Institut für Softwaretechnik und Theoretische Informatik

Area of expertise:
34351300 FG Neuronale Informationsverarbeitung

Prüfungsausschuss:
No information

Kontakt

Office:
MAR 5-6

Contact person:
Obermayer, Klaus

E-mail address:
oby@ni.tu-berlin.de

Website:

https://www.tu.berlin/ni/studium-lehre/

Learning Outcomes

Participants should learn basic concepts, their theoretical foundation, and the most common algorithms used in machine learning and artificial intelligence. After completing the module, participants should understand strengths and limitations of the different paradigms, should be able to correctly and successfully apply methods and algorithms to real world problems, should be aware of performance criteria, and should be able to critically evaluate results obtained with those methods. More specifically, participants should be able to demonstrate: 1) Knowledge of theory and methods of inductive learning 2) Application to problems of regression and classification (pattern recognition) 3) Understanding regarding basic concepts of neural information processing 4) Understanding regarding theoretical foundations to develop new machine learning techniques

Content

1) Foundations of inductive learning: empirical risk minimization, structural risk minimization, Bayesian inference 2) Learning and generalisation: gradient-based optimization, overfitting-underfitting, regularisation, application to regression and classification problems 3) Artificial neural networks (connectionist neurons, multilayer perceptrons, radial basis functions, deep networks, recurrent networks) 4) Statistical learning theory and support vector machines 5) Probabilistic methods and graphical models: reasoning under uncertainty, Bayesian learning for neural networks 6) Reinforcement Learning (MDP, value iteration, policy iteration, Q-learning)

Module Components

Pflicht:

All Courses are mandatory.

Course Name	Type	Number	Cycle	Language	SWS	VZ
Machine Intelligence I	VL	0434 L 866	WiSe	No information	2
Machine Intelligence I	UE	0434 L 866	WiSe	No information	2

Workload and Credit Points

Machine Intelligence I (VL):

Workload description	Multiplier	Hours	Total
Preparation and review	15.0	2.0h	30.0h
Time of attendance	15.0	2.0h	30.0h
			60.0h(~2 LP)

Machine Intelligence I (UE):

Workload description	Multiplier	Hours	Total
Preparation and review	15.0	6.0h	90.0h
Time of attendance	15.0	2.0h	30.0h
			120.0h(~4 LP)

The Workload of the module sums up to 180.0 Hours. Therefore the module contains 6 Credits.

Description of Teaching and Learning Methods

Lecture: Teaching in front of the class to convey the content. Exercise: Discussion of exercises which cover the mathematical derivation and analysis of neuronal methods as well as the implementation and practical usage of these methods.

Requirements for participation and examination

Desirable prerequisites for participation in the courses:

Solid Mathematical knowledge (linear algebra, analysis, and probability calculus or statistics; on a level comparable to mathematics courses for engineers) Basic programming skills (Python, Matlab, R, or Julia) Good command of the English language

Mandatory requirements for the module test application:

This module has no requirements.

Module completion

Grading

graded

Type of exam

Written exam

Language

English

Duration/Extent

90 min.

Duration of the Module

The following number of semesters is estimated for taking and completing the module:
1 Semester.

This module may be commenced in the following semesters:
Wintersemester.

Maximum Number of Participants

The maximum capacity of students is 200.

Registration Procedures

The registration for the written exam is possible at the end of the term through the electronic system of TU Berlin (as of 2017: QISPOS) or alternatively in written form via the examination office. The written exam is held in English. Both potential re-examinations are oral exams. Other than that, the examination procedure is regulated by the General Examination Regulation of the TU Berlin (AllgStuPO) and by the Examination Regulation of the Master Program Computational Neuroscience. Further information regarding registration and course material are available via the respectively current ISIS course.

Recommended literature
No recommended literature given

Assigned Degree Programs

Semester:

Zeige auch ausgelaufene Studiengänge und StuPOs

This module is used in the following Degree Programs (new System):

	Studiengang / StuPO	StuPOs	Verwendungen	Erste Verwendung	Letzte Verwendung
This module is not used in any degree program.

Students of other degrees can participate in this module without capacity testing.

Miscellaneous

No information

Machine Intelligence I