ANONYMOUS wrote Please see the announcement that just went out.

ANONYMOUS wrote The official instructions are as given on your exam timetable and the sample exam. Neither of these specify a limit.

ANONYMOUS wrote I have been desperately trying to get your marks done and released before the exam, but I have had to take some urgent leave part of yesterday and today, and have only just been able to find someone else to take over the remaining mar...

ANONYMOUS wrote There is not a clear answer, since as I say In a system with no known model where we are not able to assume anything about the structure of the system, they become basically equivalent. This information is covered in the lectures. In g...

As was discussed in the lectures The upside of Q-learning is that it learns the transition model as part of the (state, action) utilities (where regular learning requires us to either know the transition model or learn it separately). The primary down...

ANONYMOUS wrote I am not responsible for exam scheduling. That is handled entirely by the central examinations team and you would have to reach out to them with any questions. Cheers, Gozz

"Merit" is a deliberately general term that refers to anything good or positive about the technique in question. Basically you should be talking about and comparing what makes different techniques good and why you would want to use them. Gozz

ANONYMOUS wrote The latter. This is in the marks for the report, not the agent. I want to see evidence that you actually thought about the problem. If someone claims that they just came up with a single idea, implemented it, and they think it works g...

Just be honest. You investigated that idea, but were not able to get it working. So you are unable to evaluate it, but you suspect it would perform better if you had been able to implement it. That's fine. Your report should be evidence of your proble...

Please see the previous thread about this, and my response there https secure.csse.uwa.edu.au run help3011?p np opt U55

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, im...

ANONYMOUS wrote Good catch. Yes it seems that the person(s) who developed this code changed their mind at some point and forgot to change the documentation. This is why we provide testing code, however, so that you are able to make sure your agent ac...

ANONYMOUS wrote The full statement from the description is "If it is illegibly formatted (tiny font, for example) or otherwise unintelligible, it may not receive a mark." I do not believe anyone could honestly claim that 11-point font on an A4 page is...

Remember that the testing code provided is just a simple tournament boiling it down to a single empirically-measured number. In this case, what it seems really happened is that the SatisfactoryAgent lucked out and got an anomalously high proportion of...

ANONYMOUS wrote The measured win rate converges to the expected win rate as more samples are taken. The expected win rate is the intrinsic true value to which it converges. We will take a reasonable statistical sample. If your agent learns more an it...

ANONYMOUS wrote Please see the highlighted announcement on this help forum titled "Project Correction and Extension", posted Tuesday 2024-10-01.

ANONYMOUS wrote As has been addressed before, your agent should not assume it remains between games. To be considered to consistently outperform another agent, then your agent, when dropped into an identical random situation as the other agent, shoul...

Yes. All games in simulation will adhere to the rules of the game as specified, including that it says "5-10 players". Team sizes and required betrayals have not even been specified for outside this range.

ANONYMOUS wrote I am not intending to provide any fixed rules about what combinations will be tested. Note that assessing your own agent's performance is explicitly part of this project. Given that your agent is not receiving any different treatment ...

ANONYMOUS wrote Definitely looks like an interesting article, but since it is focused on deep learning, I want to make it clear that that is not a part of the content of this unit. This article is probably extremely helpful to anyone doing CITS5017 D...

ANONYMOUS wrote We do intend to run more samples when assessing your submissions, but 1000 games is still a lot, so if your agent is not consistently winning with 1000 games, it is probably a close enough thing that it's not "consistent". So submittin...

ANONYMOUS wrote No, you should not assume that your agent will survive between games. The testing code provided is not exactly the same as how your code will be assessed, and game outcome is provided so your agents as you are developing them can kn...

ANONYMOUS wrote Short answer This implementation detail is irrelevant to your task, and you should just ignore it. Long answer I did not write this code, but based on what it is doing this is how it is picking the agents to use in each game. A game c...

ANONYMOUS wrote Absolutely. Some students seem to have hyper-focused on the warning in the project and misinterpreted it. It says WARNING The reference agents BasicAgent and SatisfactoryAgent have been provided to enable you to assess your code, but ...

ANONYMOUS wrote Your agent will be tested (as in the tournament code) against various combinations of agents. It is not valid to assume that any particular agent (or any agent at all) will play randomly. Furthermore, if you were to do so, this would s...

ANONYMOUS wrote You are explicitly required to subclass the Agent class from agent.py . My previous statements have advised students not to analyse the benchmark agents provided (BasicAgent, SatisfactoryAgent), as you should be solving the problem yo...

Hi, For future reference. Please make a new thread for a new topic such as this, rather than extending an old one. Your agent will play games as both spy and loyalist. Your agent's win rate is what is being assessed. It would not make any sense for an ...

ANONYMOUS wrote You are good to start now. As the project description says you are expected to research the problem and develop a solution. This is a third-level unit with relatively open-ended assessment. You likely will be able to reasonably usin...

ANONYMOUS wrote The testing code only supports Unix systems, not Windows. If you do not yet already have a working Unix setup, I would recommend that you use Windows Subsystem for Linux (WSL) for its simplicity of setup on Windows. This is because in ...

I encourage you to begin by playing some games or watching recordings of people playing to make sure you are comfortable with the rules and mechanics. A quick search of youtube surfaced this (https youtu.be g QRczGzXqw)

ANONYMOUS wrote Hi, Per the tutorial and the warning in the project spec, there is no reason for you to read or understand BasicAgent and SatisfactoryAgent, and I (softly) discourage you from doing so. Specifically, you are implicitly and explicitly fo...

I expect it to only require English explanation. It can't hurt and may be advisable to be comfortable writing pseudocode if you need it to explain your solution, but I would say no question requires it.

ANONYMOUS wrote Details of the Mid-Semester test have now been released on LMS

ANONYMOUS wrote I hope to release information on the Mid-Semester Test and sample questions in the lecture this week (exactly two weeks before the test), but if I am not able to finalize some information by then I will release it as soon as possible ...

Tim has kindly re-recorded his lecture and uploaded it.

Another minor error The printer transposes the map along the diagonal. It should be k (c, 9 - r) .

SPOILERS for Lab 1 below In the tutorial on 2024-08-01, we discussed various approaches to solving Lab 1 (MazeAgent), and implemented some of these as we worked our way to a solution. Unfortunately, and in traditional form for live coding, my final im...