When you say we are meant to assess the performance of our agent, do you mean we have to create tests ourselves as well, or are we also meant to create several different agents and compare it to ours, and provide graphs,/stats etc in the report? IS this what you had in mind when saying we need to assess our own agent? Thanks!

ANONYMOUS wrote: The requirement is deliberately non-specific. The objective of your agent is clear: To win the game as much as possible. To do a good job of investigating this problem and developing an agent to solve it, you can't just take a guess, implement the first thing you think of, and it's good. You will have to assess if it is good. Since this is meant to be simulating encountering a real world problem that needs solving, there is no one to tell you how to assess your agent. You have to figure out how to effectively assess your agent yourself. Ultimately, your job is to convince yourself (and me) that your agent is effective, and you have to figure out how to assess it to determine that. Cheers, Gozz