Overview

Proseminar: Intelligent Internet of Things

Study Program Level: B.Sc.
Content:
This proseminar involves searching literates, learning new concepts, and developing new ideas on the crossroad of the internet of things and machine learning/artificial intelligence. Each student chooses one topic from the given list. He/she does a literature search on the selected topic and summarizes the understandings in a report (eight pages). Finally, the student holds a presentation (approximately 20 minutes) to inform other participants about the research. Active participation in the sessions is necessary.
The projects' topics belong to the following areas (All in connection with artificial intelligence and machine learning):

  1. Basics of IoT
  2. IoT Architecture
  3. IoT Connectivity
  4. IoT Enabling Technologies
  5. IoT Challenges
  6. IoT Applications

Qualification Goals:
After the lectures, the students have basic knowledge of the application field Internet of Things, with an emphasis on its intersection with artificial intelligence and machine learning. Besides, they can independently explore scientific papers concerning a specific topic, select the most relevant ones, study, and summarize the findings in a written report. Moreover, they can present, criticize, and discuss cutting-edge research scientifically.
Type: Proseminar
Frequency: Every Winter Semester
Time and Location of Introductory Session:
Tuesday 18.10.2022, 10:00-12:00, (Maria-von-Linden-Str. 6; Lecture hall on the ground floor)
Credit: 3 ECTS
Registration: via ILIAS
Links:

Important Remarks:
1. This proseminar has a limited capacity. Registration follows a first-come-first-served basis, which is managed automatically by the queuing system of ILIAS.
2. The language of instruction is English. The students can hold presentations and write the report in German.
3. It is recommended to have basic knowledge of machine learning (e.g., INF3151) and communication networks (e.g., INF3331).
4. The proseminar is graded based on the quality of the final presentation (20 minutes) and the report (eight pages).

Back to overview


Proseminar: Ambient Intelligence

Study Program Level: B.Sc.
Content:
This proseminar involves searching literates, learning new concepts, and developing new ideas on the crossroad of sensing, processing, and communication with machine learning/artificial intelligence. The objective is to become familiar with the methods to create smart environments, both on the hardware and software levels, together with the potentials and pitfalls. Each student chooses one topic from the given list. He/she does a literature search on the selected topic and summarizes the understandings in a report (approximately eight pages). Finally, the student holds a presentation (approximately 20 minutes) to inform other participants about the research. Active participation in the sessions is necessary.
The projects' topics belong to the following areas (All in connection with artificial intelligence and machine learning):
•    Situational Awareness
•    Internet of Things
•    Sensor Networks
•    Smart Environments
•    Human-Computer Interaction
•    Assistive Environment
•    Agents and Ambient Intelligence
•    Pervasive and Mobile Computing
•    Edge Computing
•    Wearable Computing
Qualification Goals:
After the lectures, the students have basic knowledge of the application field Ambient Intelligence, with an emphasis on the intersection between sensing, processing, and communication with artificial intelligence and machine learning. Besides, they can independently explore scientific papers concerning a specific topic, select the most relevant ones, study, and summarize the findings in a written report. Moreover, they can present, criticize, and discuss cutting-edge research scientifically.
Type: Proseminar
Frequency: Every Summer Semester
Credit: 3 ECTS
Registration: via ILIAS
Links:
ILIAS Page
ALMA Page
Important Remarks:
1. This proseminar has a limited capacity. Registration follows a first-come-first-served basis, which is managed automatically by the queuing system of ILIAS.
2. The language of instruction is English. The students can hold presentations and write the report in German.
3. It is recommended to have basic knowledge of machine learning (e.g., INF3151) and communication networks (e.g., INF3331).
4. The proseminar is graded based on the quality of the final presentation (20 minutes) and the report (eight pages).

Back to overview

 


Seminar: Multi-Agent Reinforcement Learning

Study Program Level: M.Sc.
Content:
This seminar involves searching literates, learning new concepts, and developing new ideas on the crossroad of multi-agent systems and game theory with machine learning and artificial intelligence. Each student chooses one topic from the given list. He/she does a literature search on the selected topic and summarizes the understandings in a report (approximately eight pages). Finally, the student holds a presentation (approximately 20 minutes) to inform other participants about the research. Active participation in the sessions is necessary. The report and talk shall include at least three papers. The students can work in team consisting of at most two persons. In this case, the report and talk shall include at least six papers.
The projects' topics belong to the following list (related areas and cross-topic papers are valid as well):
•    Cooperative MA-RL
•    Deep MA-RL
•    Competitive MA-RL
•    Mean-field MA-RL
•    Hierarchical MA-RL
•    Multi-objective MA-RL
•    Off-policy MA-RL
•    MA-RL with transfer learning
Qualification Goals:
After the seminar, the students have a broad knowledge over the research direction in multi-agent reinforcement learning. Besides, they can independently explore scientific papers concerning a specific topic, select the most relevant ones, study, and summarize the findings in a written report. Moreover, they can present, criticize, and discuss cutting-edge research scientifically.
Type: Seminar
Frequency: Every Summer Semester
Credit: 3 ECTS
Registration: via ILIAS
Links:
ILIAS Page
ALMA Page
Important Remarks:
1. This seminar has a limited capacity. Registration follows a first-come-first-served basis, which is managed automatically by the queuing system of ILIAS.
2. The language of instruction is English. The students can hold presentations and write the report in German.
3. It is recommended to have basic knowledge of machine learning and game theory.
4. The seminar is graded based on the quality of the final presentation (20 minutes) and the report (six pages).

Back to overview