|
|
|
Problem Set 2 Due
Wednesday 10/22 at 1:30pm (Hand in printout at start of class)
See the policy on lateness and collaboration.
In all problems below, please think carefully before answering questions – strive for a terse, precise explanation. Points may be taken off for overly long or complex explanations. If it is impossible answer a question with the information given, feel free to make additional assumptions (just state them clearly).
1. R&N – exercise
7.9
2. R&N – exercise 8.6
a-d
3. R&N – exercise
8.7
4. R&N – exercise
9.4
5. Start planning your
project.
a. Choose a name for your
team or for the agent you aspire to create.
b. What is the name of the
classmate who will fill out your team. (teams should be 2 people – other sizes require advance
permission).
In a separately turned in document (one per team) please explain your
plan for the project. Each plan may be brief (at this point) e.g. ~2 pages, but
should include (at least) the following:
1. Team name
2. Team members
3. Overall architecture for
agent
4. What kind of state your
agent will keep (e.g. enemy location, model of enemy behavior, no state, ??) during a battle?
5. What heuristic function(s)
will you use and how will you compute them from the robocode
statistics?
6. What operator(s) will you
use to modify agents?
7. What kind of search
strategy will you employ?
Computer Science & Engineering
Department
University of Washington
PO Box 352350
Seattle, WA 98195-2350
USA
Seattle, WA