Moses - Advanced topics in Reinforcement Learning

Display language

To modulepage Generate PDF

Module / Version:
#40969 / #2

Validity:
Seit WiSe 2022/23
(Deaktivierung beantragt zum SoSe 2024)

Default display language:
English

Module title:
Advanced topics in Reinforcement Learning

Credits:
6

Responsible person:
Obermayer, Klaus

Grading:
benotet

Type of exam:
Portfolioprüfung

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:
Groiß, Camilla

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

Website:

http://www.ni.tu-berlin.de/teaching/

Learning Outcomes

After successful completion, participants are able to apply at least one state-of-the-art technique in Deep Reinforcement Learning and present their results to other students. They will be able to reproduce and communicate scientific research and also have a more specific and deeper knowledge of how the brain works depending on reinforcement.

Content

Participants will survey the recent literature on Reinforcement Learning, present one state-of-the-art method of their choice and then apply it in a coding project.

Module Components

Pflichtgruppe:

All Courses are mandatory.

Course Name	Type	Number	Cycle	Language	SWS	VZ
Advanced topics in Reinforcement Learning	SEM	3435 L 10262	WiSe/SoSe	English	2

Workload and Credit Points

Advanced topics in Reinforcement Learning (SEM):

Workload description	Multiplier	Hours	Total
Attendance	15.0	2.0h	30.0h
Pre/post processing	15.0	10.0h	150.0h
			180.0h(~6 LP)

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

Description of Teaching and Learning Methods

Participants will research the relevant literature on Reinforcement learning, more specifically in topics related to safety and control. Each student will have at least one chance to present a paper. Students will carry out a coding project and complete an individual research project or an in-depth literature survey at the end of the course. The module will be held in English.

Requirements for participation and examination

Desirable prerequisites for participation in the courses:

Required skills include: literature research, presentation of researched articles; students should already be familiar with artificial intelligence Participants are expected to have background knowledge comparable to the topics covered in Machine Intelligence I + II .

Mandatory requirements for the module test application:

This module has no requirements.

Module completion

Grading

graded

Type of exam

Portfolio examination

Type of portfolio examination

100 Punkte insgesamt

Language

English

Test elements

Name	Points	Categorie	Duration/Extent
(Deliverable assessment) Coding project	30	practical	30 hours
(Deliverable assessment) Project presentation	50	practical	40 minutes
(Deliverable assessment) Literature review	20	written	6 hours

Grading scale

Notenschlüssel »Notenschlüssel 2: Fak IV (2)«

Gesamtpunktzahl	1.0	1.3	1.7	2.0	2.3	2.7	3.0	3.3	3.7	4.0
100.0pt	95.0pt	90.0pt	85.0pt	80.0pt	75.0pt	70.0pt	65.0pt	60.0pt	55.0pt	50.0pt

Test description (Module completion)

Portfolio exam 100 points in total

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:
Winter- und Sommersemester.

Maximum Number of Participants

The maximum capacity of students is 30.

Registration Procedures

registration via MOSES; exam takes place in English language

Recommended literature
Sutton, R.S. & Barto, A.G.: reinforcement learning: An introduction. MIT press, 2018
Bertsekas, D.: Reinforcment Learning and Optimal Control. Athena Scientific, 2019
Amodei, D., Olah, J., Steinhardt, J. Christiano, P., Schulman, J. and Mane, D.: Concrete Problems in AI Safety. 2016

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
Computer Engineering (M. Sc.)	1	12	WiSe 2022/23	SoSe 2024
Computer Science (Informatik) (M. Sc.)	1	12	WiSe 2022/23	SoSe 2024
Elektrotechnik (M. Sc.)	1	8	WiSe 2022/23	SoSe 2024
Information Systems Management (Wirtschaftsinformatik) (M. Sc.)	1	4	WiSe 2022/23	SoSe 2024

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

Miscellaneous

No information

Advanced topics in Reinforcement Learning