Physics Department, TUM | Course 0000002676 in SS 2023

General Data

Course Type

practical training

Semester Weekly Hours

6 SWS

Organisational Unit

Informatics 2 - Chair of Formal Languages, Compiler Construction, Software Construction (Prof. Seidl)

Lecturers

Tobias Eder
Responsible/Coordination: Georg Groh

Dates

Wed, 16:00–18:00, MI 01.07.023
and 1 singular or moved dates

iCalendar

	Date and Time	Room and Remark
!	Wed, 2023-04-19, 17:00–19:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc. First session one week earlier than originally planned because of a business trip the following week.
	Wed, 2023-05-17, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.
	Wed, 2023-05-24, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.
	Wed, 2023-05-31, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.
	Wed, 2023-06-07, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.
	Wed, 2023-06-14, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.
	Wed, 2023-06-21, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.
	Wed, 2023-06-28, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.
	Wed, 2023-07-05, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.
	Wed, 2023-07-12, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.
	Wed, 2023-07-19, 16:00–18:00	Boltzmannstr. 3 Speaker: Tobias Eder M.Sc.

Assignment to Modules

IN2106: Master-Praktikum / Advanced Practical Course
This module is included in the following catalogs:
- Further Modules from Other Disciplines

Further Information

Courses are together with exams the building blocks for modules. Please keep in mind that information on the contents, learning outcomes and, especially examination conditions are given on the module level only – see section "Assignment to Modules" above.

additional remarks	With the developments of AI in the past 10 years the applications for such systems have grown exponentially. There is a huge need to apply AI-based systems both commercially and practically for dealing with large volumes of data today and to manage and drive decision making in a myriad of fields. At the same time, increasingly complex learning systems have led to a methodological smoke screen effect that makes it hard to link intention to actual effect in some applications. The potential for both intentional and unintentional harm can be large and is a serious problem in acceptance of these techniques in critical decision-making processes. This lab course will focus on designing and applying ethical concepts and constraints to data-driven AI applications. This course therefore aims to both operate at the state-of-the-art in AI and ML and see what problems emerge when using known data-driven approaches and how to combat them. In projects teams are tasked to work on application-oriented projects to deal with known problems such as Bias, Fairness, Privacy, Trust, Transparency, and the potential for abuse with both discriminative and generative models. The lab will offer different projects focused on specific problems, with a stronger focus on Natural Language Processing systems. There will however also be topics without a strict NLP focus and some for which the field of application can be freely chosen by the group members. Matching procedure via matching.in.tum.de. Preliminary Meeting Wednesday, Feb 1, 16:00 in room 01.07.014 Garching FMI - Deregistration deadline: 15 March 2021 Please apply to the course beforehand at https://wiki.tum.de/display/socialcomputing/Ethical+AI%3A+Problems+and+Applications+SS2023 - you can also find the slides of the preliminary meeting there.
Links	TUMonline entry

additional remarks

With the developments of AI in the past 10 years the applications for such systems have grown exponentially. There is a huge need to apply AI-based systems both commercially and practically for dealing with large volumes of data today and to manage and drive decision making in a myriad of fields. At the same time, increasingly complex learning systems have led to a methodological smoke screen effect that makes it hard to link intention to actual effect in some applications. The potential for both intentional and unintentional harm can be large and is a serious problem in acceptance of these techniques in critical decision-making processes. This lab course will focus on designing and applying ethical concepts and constraints to data-driven AI applications. This course therefore aims to both operate at the state-of-the-art in AI and ML and see what problems emerge when using known data-driven approaches and how to combat them. In projects teams are tasked to work on application-oriented projects to deal with known problems such as Bias, Fairness, Privacy, Trust, Transparency, and the potential for abuse with both discriminative and generative models. The lab will offer different projects focused on specific problems, with a stronger focus on Natural Language Processing systems. There will however also be topics without a strict NLP focus and some for which the field of application can be freely chosen by the group members. Matching procedure via matching.in.tum.de. Preliminary Meeting Wednesday, Feb 1, 16:00 in room 01.07.014 Garching FMI - Deregistration deadline: 15 March 2021 Please apply to the course beforehand at https://wiki.tum.de/display/socialcomputing/Ethical+AI%3A+Problems+and+Applications+SS2023 - you can also find the slides of the preliminary meeting there.

Links

TUMonline entry

Equivalent Courses (e. g. in other semesters)

Semester	Title	Lecturers	Dates
SS 2024	Master Lab Course – Ethical AI: Problems and Applications (IN2106, IN4297)	Eder, T. Responsible/Coordination: Groh, G.	Wed, 16:00–18:00, MI 02.07.014 and singular or moved dates
SS 2022	Master Lab Course – Ethical AI: Problems and Applications (IN2106, IN4297)	Eder, T. Responsible/Coordination: Groh, G.	Wed, 16:00–18:00, MI 00.08.055 and singular or moved dates
SS 2021	Master Lab Course – Ethical AI: Problems and Applications (IN2106, IN4297)	Eder, T. Responsible/Coordination: Groh, G.	singular or moved dates

Master Lab Course – Ethical AI: Problems and Applications (IN2106, IN4297)

Course 0000002676 in SS 2023

General Data

Assignment to Modules

Further Information

Equivalent Courses (e. g. in other semesters)