|Period:||kan in elke periode op individuele basis|
|Participants:||up till now 9 subscriptions|
|Schedule:||Note: from now on the schedule is to be found in Osiris|
|Nota bene:||Er is geen recente vakbeschrijving beschikbaar.|
Onderstaande tekst is een oude vakbeschrijving uit collegejaar 2010/2011
|Contents:||The goal of an experimentation project is to let a student experience what it means to set up and conduct an experiment to test a hypothesis. The project will consist of the following steps:
The project should be done individual or in a group of 2 students and should be supervised by one of the staff members teaching in the AT masters. The proposal for the project has to be approved by the coordinator and the supervisor before any practical work can be started.
- Setting up a hypothesis that can be tested. Of course the hypothesis should relate to a topic in the AT masters. The following are examples of possible experimentation projects:
- The performance of the 2APL interpreter will be improved if its rule selection mechanism is based on a learning mechanism, an argumentation framework, or a specific theorem prover.
- Agents with specific behaviours following a specific interaction protocol can (or cannot) achieve a global objective.
- The second step is to device a test with which this hypothesis can be verified.
- Next, an implementation should be made in order to run an experiment. For example, an implementation of a new rule selection mechanism for 2APL should be implemented (or a simulation should be implemented to show that a group of agents with specific behaviour following a specific interaction protocol can achieve a global objective).
- The experiment is conducted on the implemented software.
- An analysis is made of the experiment to check whether the hypothesis is supported or not.
|Minimum effort to qualify for 2nd chance exam:||Om aan de aanvullende toets te mogen meedoen moet de oorspronkelijke uitslag minstens 4 zijn.|