Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
Assignment 2 requires that you test existing controllers and develop and improve your own controller based on techniques covered during this course. We only consider the Single Player Planning Track in this assignment. For a quick start, take a look at src/tracks/singlePlayer/Test.java for an example on how to use the framework.
In this assignment, we only consider games from Training Set 1. For each exercise, you have to evaluate the considered controllers on the following 4 games: Aliens, Boulderdash, Butter ies, Chase. Note that each game includes 5 levels (from \lvl0″ to \lvl4″), so you will be looking at 20 game levels in total.
Just like in Assignment 1, we’d like each team member to write one paragraph about what he or she has contributed to this assignment. We will not mark this, and it will not have any eect on the marking of the other exercises. You might now ask, \why do this then?” { well, through this no-stakes approach, we’d like to encourage self-regulation within the group and cooperative learning. You can’t lose; you can only win.
1. Run each of the three controllers 10 times on each of the 5 levels of each game (20 levels in total) and report for each game level the average score and standard deviation.
2. Change the \One Step Look-Ahead” Controller such that it looks more than 1 step ahead. Make sure that it does not exceed the time out and try to obtain a high performing controller.
3. Run your new \several steps ahead”-controller on each game 5 times and report the average score and standard deviation. In the report, you must brie y describe and justify the changes that you have made.
Consider the \Genetic Algorithm” Controller
1. Design an evolutionary algorithm to optimise the parameter setting (GAMMA,SIMULATION DEPTH, POPULATION SIZE, RECPROB, MUT) for the \Ge-netic Algorithm” Controller for all levels of a considered game (5 levels). The objective is to maximise the scores this controller achieves on these games. In the report, you must brie y describe the algorithm and justify its design.
2. Give the best obtained parameter settings for each of the 4 games together with the results obtained for each game and compare it to the default parameter setting tested in Exercise 2. The comparison should be statistically meaningful (i.e. using statistical testing methods).