Open the "task" files to see what you need to do.
Write your findings into the "journal" files.
You will receive points between 1 and 10 for each completed exercise.
Grading criteria include:
- Was anything submitted at all?
- Did you document what you did in the lab journal?
- Did you test different ideas? What did you find?
- How effective is the submitted behavior? Does the agent achieve a high utility? How well does it in comparison to the others?
- How original is the submitted behavior? Does your agent apply heuristics that are unlike those of the other agents?
- How adaptive is the submitted behavior? Does your agent still perform well if the parameters of the simulation are adjusted?
- How clean and well-documented is your source-code?