Department of Computer Science Technische Universitaet Muenchen
Chair XVI: Computer Aided Medical Procedures & Augmented Reality Home | Teaching | People | Publications | Labs & Locations| Applications | Open Positions | History | Partners | Internal
MQM  Model-Based Systems & Qualitative Reasoning            
Home | Projects | Teaching | People | Publications | Theses & Guided Research | Jobs

Seminar: Automated Problem Solving in Artificial Intelligence

Content:

Automating problem solving, especially solving technical problems, such as diagnosing a device, generating tests, and configuring a system is one major aim of artificial intelligence. Model-based systems represent a class of knowledge based systems that are currently employed in more and more industrial areas. Their speciality is the distinct separation of the system represented by a model and the task specific problem solving knowledge. By separating system model and problem solving techniques it becomes possible to use the model for different problem settings.

Example Applications:

  • Fault Diagnosis in vehicles and other technical systems.
  • Environment Modeling and Decision Support Systems.
  • Current research at MQM: Automatic Training plan generation and interactive support for trainers and trainees in fitness studios.
  • Supervisor:

    Prof. Peter Struss, Florian Grigoleit, and Robert Reiser.
    For questions, please contact Florian Grigoleit

    Aim:

    The students learn in small teams to acquaint themselves in a scientific topic, to analyze literature, to write a scientific paper, and to present their results in front of academic peers.

    Modules: IN0014, IN2107, IN4826
    Miscellaneous: Number of students is limited to 14
    Time and Room: TBA
    Start: 17.10.2016
    Language: English
    Preliminary Discussion (Vorbesprechung) : 6.7.2016 10:30 00.13.36

    36 Process:

  • Students will work in small teams (2 students / team).
  • The results have to be presented in English. The presentations should be about 20 minutes per student
  • The students write a 8 to 10 page paper (English / German) on their topic. A Word-Template for the paper will be provided.
  • Two paper critiques of 1 to 2 pages must be created by every student.
  • Deadlines:

  • Presentation Slides (.ppt / .pptx): One week before the presentation date
  • Paper Draft Submission(.doc / .docx): 31.1.2017
  • Final Paper Submission(.doc / .docx): 28.2.2017

  • Topics:

    The student teams read and analyze relevant research articles. For the paper critiques one application and one theoretical paper have to be analyzed and assessed. The student paper must provide an introduction to the topic, a description of the basic principles and techniques, followed by the analysis of an application and of one current research topic with at least one modern solution approach.

  • Knowledge-based Configuration
    • Topic: Rule-based Systems - R1
    • Advisor: Florian Grigoleit

  • Constraint-based Problem Solving
    • Topic: Analysis of i) constraint types and ii) reasoning techniques
    • Advisor: Florian Grigoleit

  • Knowledge-Representation Techniques
    • Topic: Representing knowledge for problem solvers
    • Advisor: Florian Grigoleit

  • Truth Maintenance Systems
    • Topic: Types of TMSs and connection with problem solvers
    • Advisor: Florian Grigoleit

  • Model-based Problem Solving
    • Topic: Consistency-based reasoning
    • Advisor: Robert Reiser

  • Qualitative Modelling
    • Topic: Applications, reasoning techniques, limits, and current topics, component and process oriented modelling. Challenges in qualitative modelling (intervals, abstraction, etc.)
    • Advisor: Robert Reiser

  • Model-based Diagnosis
    • Topic: Principles, applications, algorithms, current topics
    • Advisor: Robert Reiser

     




    Computer Science XVI | Technische Universität München