Advanced methods of artificial intelligence 1000-2M12ZMI
In the recent years one of the most spectacular triumphs of AI was winning an American TV show Jeopardy by Watson, computer created by IBM. It defeated two former human champions in tasks requiring versatile analysis of texts formulated in natural language. During the show Watson had at his disposal over 200 million pages and full text of Wikipedia, but not the Internet connection. This impressive result was possible due to synergy of specialized algorithms and a variety of AI techniques. During the lecture we will show, how the synthesis of these methods allowed Watson to win the prestigious TV show. To do that we will introduce a number of AI techniques, including:
- rule systems
- Rete networks
- fuzzy systems
- planning methods
- autonomic processing of events.
Crucial requirement for modern autonomous systems is getting a realistic model of the environment. We will show a new approach to this problem, taking into consideration both incomplete and inconsistent information. Additionally, the lecture will be illustrated by AI applications such as warfare simulation and social network analysis.
During the laboratories we will create applications using different AI techniques implemented in the system ABLE: Agent Building and Learning Environment. ABLE was created by IBM as a Java-based open environment using the same methods as commercial agent systems developed by major companies. It contains a rich set of AI algorithms for different types of reasoning and planning. Additionally, ABLE contains a variety of components and tools simplifying the design and implementation of intelligent agents.
Type of course
Assessment criteria
Each student’s final grade/assessment consists of two mandatory parts:
Presenting a selected subject during classes, prepared basing on
listed course material and recommended references (30% of the final
assesment).
Designing and implementing a specific case study in ABLE during LAB.
(70% of the final assesment)
Additional information
Information on level of this course, year of study and semester when the course unit is delivered, types and amount of class hours - can be found in course structure diagrams of apropriate study programmes. This course is related to the following study programmes:
- Bachelor's degree, first cycle programme, Computer Science
- Master's degree, second cycle programme, Computer Science
Additional information (registration calendar, class conductors, localization and schedules of classes), might be available in the USOSweb system: