Intelligent systems
Overall Course Objectives
This course aims to provide participants with the fundamental competences required to conceive, design and implement intelligent and distributed control systems combining data-driven and knowledge-based techniques. The course work offers participants an application-oriented introduction to machine learning and declarative methods of artificial intelligence. Course participants will learn the theoretical aspect of intelligent systems with respect to use-cases and top level design to design.
See course description in Danish
Learning Objectives
- Identify a real-world problem and propose an intelligent system that can be used to help with this problem.
- Identify and discuss situation awareness, decision, and control tasks in an intelligent system and identify the hardware and/or software requirements.
- Discuss whether machine learning methods are applicable for a proposed intelligent system, and (if relevant) propose a suitable machine learning algorithm/method and evaluate its effectiveness.
- Work with a large data set, perform data exploration and feature extraction to derive knowledge from data in a real-world automation problem.
- Explain structured knowledge representation techniques and related modeling principles, and select suitable knowledge representations for integration in an intelligent systems solution.
- Describe suitable uses of logic-based methods applied to the solution of decision making problems in intelligent systems.
- Explain and select appropriate methods as well as the corresponding data models and knowledge representations as part of a design problem.
- Analyse an intelligent systems design problem, communicate the design idea, formulate specifications and test requirements.
- Design operational prototype of an intelligent system using data-driven techniques applied to a distributed systems problem.
- Write a project proposal for an intelligent system that is intended to solve a real-world problem and discuss how the functionality of the implementation can be tested and verified and how critical parts/elements can be identified.
- Discuss the quality and availability of data for an intelligent system and how GDPR rules can affect the implementation or operation of an intelligent system.
- Give appropriate feedback on a project proposal
Course Content
Introduction to intelligent systems problems, task analysis, systems development approaches & architecture elements; applications of intelligent systems in automation, Internet of Things (IoT) and Smart Grid / Energy problems.
Distributed systems basics: concurrency, communication, and debugging.
Structured knowledge representation (e.g. ontologies) and applications to declarative and logic-based programming methods (e.g. rule-based logic, graph-search, … ).
Handling large data sets; data visualization; applied statistical learning methods. Quality of data-driven models.
Recommended prerequisites
Experience with script programming (e.g. MATLAB, Python, etc.) and object-oriented programming (e.g. Java, C++, or python etc.).
Teaching Method
Lectures, exercises and group work. Project planning work i groups
Faculty
Limited number of seats
Minimum: 6.
Please be aware that this course will only be held if the required minimum number of participants is met. You will be informed 8 days before the start of the course, whether the course will be held.