PLEASE NOTE: this server, secure.csse.uwa.edu.au, will be unavailable on Monday 5th December, from 1pm until 4pm.
Web-based programs, such as csmarks, cssubmit, and the help fora, will all be unavailable.
  It's UWAweek 49

help3001

This forum is provided to promote discussion amongst students enrolled in CITS3001 Algorithms, Agents and Artificial Intelligence.

Please consider offering answers and suggestions to help other students! And if you fix a problem by following a suggestion here, it would be great if other interested students could see a short "Great, fixed it!"  followup message.

How do I ask a good question?
Displaying the 3 articles in this topic
Showing 3 of 175 articles.
Currently 6 other people reading this forum.


 UWA week 35 (2nd semester, week 6) ↓
SVG not supported

Login to reply
4:13pm Sun 4th Sep, Tin P.

Hey, I've got quite a few questions.

  1. Can blue agents lose followers?

    • Red can lose followers, but it's not stated if blue can also lose followers.
  2. Do Grey agents have turn actions? Or do they act like Green agents once introduced?

    • When Blue introduces a Grey agent that's not a spy, I'm assuming it will be a part of the Green population and interact with other Greens when it's Green's turn to influence each other. Blue will also interact with Grey as if it's a Green, with the only difference being that no energy cost is incurred.
      • would this Grey agent interact with all Greens? Or does it have a random chance of interaction with each Greens (specified by input probability)?
    • when Grey is a Red Spy, do they act like a second Red agent?
      • i.e. they go through the same steps that the original red agent went through in influencing each Green agent and choosing a message potency.
      • This could mean that in a game, we could have (let's say) 5 red agents each taking actions in one round, effectively meaning Red can disperse multiple messages to each green in ONE round.
    • If Red spies don't act like a second agent, how do they choose the message potency to spread to other greens?
  3. Do Grey agents persist after they are introduced?

    • i.e. Do Blue always have a free Grey agent to interact with that incurs no cost until the game ends, or does this no-cost effect only last 1 round, after which the Grey agent becomes a normal Green agent?
  4. It says the population initially has 90% Green and no Grey agents. I assumed that Red and Blue will each only be 1 agent, so what's the remaining 10%?

  5. Is it possible for Blue to still win after being depleted of energy?

    • Say if we define a fixed number of iterations until the game ends, Blue could run out of energy at some point leaving them unable to act. Red will still be able to act, but it's possible that once the game end, the majority of Green still has the opinion to vote.
    • Or if not, do Blue automatically lose if they run out of energy?
      • With an indefinite number of iterations, Blue WILL eventually run out of energy which means it's impossible for Blue to win.
      • Or, if the win/lose condition after Blue runs out of energy is the proportion of Green that wants to vote, Blue's strategy could be to force-end the game after the majority wants to vote by purposely running out of energy. Red can't act since the game has ended, but this means Blue could have a huge advantage over Red which goes against the point of the project.
  6. How exact should we follow the project specification? I.e. are we allowed to add additional features like:

    • Blue can replenish energy gradually,
    • setting a fixed number of iterations until the game ends,
    • adding random events during each round,
    • etc...
    • I could come up with rules/limits to the questions above if we are not following the project sheet too strictly
  7. How many potent messages can each agent choose from?

    • In the project sheet it said 5 levels of potent messaging and 10 correction messages and 10 pieces of misinformation.

uh... Apologies for writing so much, a lot of questions came up while I was reading the project sheet and other forum posts.


 UWA week 36 (2nd semester, mid-semester break) ↓
SVG not supported

Login to reply
10:25pm Sun 11th Sep, Mehwish N.

"Tin Pang" [email protected] wrote:

Hey, I've got quite a few questions.

  1. Can blue agents lose followers?

    • Red can lose followers, but it's not stated if blue can also lose followers.
  2. Do Grey agents have turn actions? Or do they act like Green agents once introduced?

    • When Blue introduces a Grey agent that's not a spy, I'm assuming it will be a part of the Green population and interact with other Greens when it's Green's turn to influence each other. Blue will also interact with Grey as if it's a Green, with the only difference being that no energy cost is incurred.
      • would this Grey agent interact with all Greens? Or does it have a random chance of interaction with each Greens (specified by input probability)?
    • when Grey is a Red Spy, do they act like a second Red agent?
      • i.e. they go through the same steps that the original red agent went through in influencing each Green agent and choosing a message potency.
      • This could mean that in a game, we could have (let's say) 5 red agents each taking actions in one round, effectively meaning Red can disperse multiple messages to each green in ONE round.
    • If Red spies don't act like a second agent, how do they choose the message potency to spread to other greens?
  3. Do Grey agents persist after they are introduced?

    • i.e. Do Blue always have a free Grey agent to interact with that incurs no cost until the game ends, or does this no-cost effect only last 1 round, after which the Grey agent becomes a normal Green agent?
  4. It says the population initially has 90% Green and no Grey agents. I assumed that Red and Blue will each only be 1 agent, so what's the remaining 10%?

  5. Is it possible for Blue to still win after being depleted of energy?

    • Say if we define a fixed number of iterations until the game ends, Blue could run out of energy at some point leaving them unable to act. Red will still be able to act, but it's possible that once the game end, the majority of Green still has the opinion to vote.
    • Or if not, do Blue automatically lose if they run out of energy?
      • With an indefinite number of iterations, Blue WILL eventually run out of energy which means it's impossible for Blue to win.
      • Or, if the win/lose condition after Blue runs out of energy is the proportion of Green that wants to vote, Blue's strategy could be to force-end the game after the majority wants to vote by purposely running out of energy. Red can't act since the game has ended, but this means Blue could have a huge advantage over Red which goes against the point of the project.
  6. How exact should we follow the project specification? I.e. are we allowed to add additional features like:

    • Blue can replenish energy gradually,
    • setting a fixed number of iterations until the game ends,
    • adding random events during each round,
    • etc...
    • I could come up with rules/limits to the questions above if we are not following the project sheet too strictly
  7. How many potent messages can each agent choose from?

    • In the project sheet it said 5 levels of potent messaging and 10 correction messages and 10 pieces of misinformation.

uh... Apologies for writing so much, a lot of questions came up while I was reading the project sheet and other forum posts.

• Why does the red team have their own uncertainty? I explained this concept in the class, giving examples of android vs apple phone users. The red team has an uncertainty variable because we need to quantify whether it can pursue a green node to change its opinion or not. And that change in opinion depends upon the uncertainty of the green node. Hence you can think of red node’s uncertainty as its “persuasive power” • How does internet literacy impact the simulation? This was background information. Ignore it or use it to create a sophisticated green network (for any bonus points) • What is the goal of the red team: o Increase uncertainty so that people do not vote. Change green nodes’ opinion to “not vote”, also increase uncertainty among voters so that they eventually change their opinion and also make the opinions of those who are not voting more certain (we will look at this in lecture 7) o Change people's opinions to vote the way they want. This game has only 2 opinions: “vote” or “not vote” • How does the uncertainty of nodes change with interactions? We will look at some examples in lecture 7. It is also for you to figure out. Do some research and suggest a smart way to change the uncertainties of green nodes • Does a blue team lose the same amount of energy for interacting with green nodes that would already vote for them, as for nodes that wouldn't? Yes. • Do the red/blue teams know the opinion and uncertainty of green nodes? They do know the opinion, but not the uncertainty. • Can the red/blue teams select the agents that they want to interact with, or do they interact with all of them? They need to interact with everyone. But if you believe your model can become more sophisticated if you let red/blue interact with a few agents, then go for it. Please justify that in your report. • Are green nodes supposed to affect other green nodes? Yes, that is the purpose of interaction among green nodes; for example, common people talking to each other and influencing each other’s opinions. • Does "grey-bad" correspond to spies? Yes

  1. Can blue agents lose followers? o No
  2. Do Grey agents have turn actions? Or do they act like Green agents once introduced? They will execute a turn once and then disappear o When Blue introduces a Grey agent that's not a spy, I'm assuming it will be a part of the Green population and interact with other Greens when it's Green's turn to influence each other. Blue will also interact with Grey as if it's a Green, with the only difference being that no energy cost is incurred.  would this Grey agent interact with all Greens? Or does it have a random chance of interaction with each Greens (specified by input probability)? All greens o when Grey is a Red Spy, do they act like a second Red agent?  i.e. they go through the same steps that the original red agent went through in influencing each Green agent and choosing a message potency. Yes, but red agent will not lose followers in this case.  This could mean that in a game, we could have (let's say) 5 red agents each taking actions in one round, effectively meaning Red can disperse multiple messages to each green in ONE round. We have only ONE red agent. o If Red spies don't act like a second agent, how do they choose the message potency to spread to other greens? They do act like a red agent. A grey agent can chose a random message
  3. Do Grey agents persist after they are introduced? o i.e. Do Blue always have a free Grey agent to interact with that incurs no cost until the game ends, or does this no-cost effect only last 1 round, after which the Grey agent becomes a normal Green agent? No, they perish
  4. It says the population initially has 90% Green and no Grey agents. I assumed that Red and Blue will each only be 1 agent, so what's the remaining 10%? That was an estimate. One red, One blue, many green and a few grey.
  5. Is it possible for Blue to still win after being depleted of energy? o Say if we define a fixed number of iterations until the game ends, Blue could run out of energy at some point leaving them unable to act. Red will still be able to act, but it's possible that once the game end, the majority of Green still has the opinion to vote. YES it is possible o Or if not, do Blue automatically lose if they run out of energy?  With an indefinite number of iterations, Blue WILL eventually run out of energy which means it's impossible for Blue to win.  Or, if the win/lose condition after Blue runs out of energy is the proportion of Green that wants to vote, Blue's strategy could be to force-end the game after the majority wants to vote by purposely running out of energy. Red can't act since the game has ended, but this means Blue could have a huge advantage over Red which goes against the point of the project. o Well game will end if blue dies and then we will see how many greens want to vote.
  6. How exact should we follow the project specification? I.e. are we allowed to add additional features like: o Blue can replenish energy gradually, POSSIBLY o setting a fixed number of iterations until the game ends, POSSIBLY o adding random events during each round, POSSIBLY o etc... o I could come up with rules/limits to the questions above if we are not following the project sheet too strictly MENTION IN YOUR REPORT
  7. How many potent messages can each agent choose from? o In the project sheet it said 5 levels of potent messaging and 10 correction messages and 10 pieces of misinformation. 10 is okay


 UWA week 37 (2nd semester, week 7) ↓
SVG not supported

Login to reply
3:53pm Thu 15th Sep, Zach M.

Can blue agents lose followers?
		No

Do Grey agents have turn actions? Or do they act like Green agents once introduced?
	They will execute a turn once and then disappear

When Blue introduces a Grey agent that's not a spy, I'm assuming it will be a part of the Green population and interact with other Greens when it's Green's turn to influence each other. Blue will also interact with Grey as if it's a Green, with the only difference being that no energy cost is incurred.  Would this Grey agent interact with all Greens? Or does it have a random chance of interaction with each Greens (specified by input probability)?
	All greens 


When Grey is a Red Spy, do they act like a second Red agent?  i.e. they go through the same steps that the original red agent went through in influencing each Green agent and choosing a message potency.
	Yes, but red agent will not lose followers in this case. 

This could mean that in a game, we could have (let's say) 5 red agents each taking actions in one round, effectively meaning Red can disperse multiple messages to each green in ONE round.
	We have only ONE red agent. 

If Red spies don't act like a second agent, how do they choose the message potency to spread to other greens?
	They do act like a red agent.  A grey agent can chose a random message

Do Grey agents persist after they are introduced? o i.e. Do Blue always have a free Grey agent to interact with that incurs no cost until the game ends, or does this no-cost effect only last 1 round, after which the Grey agent becomes a normal Green agent?
	No, they perish


It says the population initially has 90% Green and no Grey agents. I assumed that Red and Blue will each only be 1 agent, so what's the remaining 10%?
	That was an estimate. One red, One blue, many green and a few grey.

Is it possible for Blue to still win after being depleted of energy? o Say if we define a fixed number of iterations until the game ends, Blue could run out of energy at some point leaving them unable to act. Red will still be able to act, but it's possible that once the game end, the majority of Green still has the opinion to vote.
	YES it is possible

Or if not, do Blue automatically lose if they run out of energy?  With an indefinite number of iterations, Blue WILL eventually run out of energy which means it's impossible for Blue to win.  Or, if the win/lose condition after Blue runs out of energy is the proportion of Green that wants to vote, Blue's strategy could be to force-end the game after the majority wants to vote by purposely running out of energy. Red can't act since the game has ended, but this means Blue could have a huge advantage over Red which goes against the point of the project.
	Well game will end if blue dies and then we will see how many greens want to vote.


How exact should we follow the project specification? I.e. are we allowed to add additional features like: o Blue can replenish energy gradually
	POSSIBLY o setting a fixed number of iterations until the game ends
	POSSIBLY o adding random events during each round
	POSSIBLY o etc...
	I could come up with rules/limits to the questions above if we are not following the project sheet too strictly MENTION IN YOUR REPORT


 How many potent messages can each agent choose from? o In the project sheet it said 5 levels of potent messaging and 10 correction messages and 10 pieces of misinformation.
	10 is okay

The University of Western Australia

Computer Science and Software Engineering

CRICOS Code: 00126G
Written by [email protected]
Powered by history
Feedback always welcome - it makes our software better!
Last modified  1:17AM Sep 14 2022
Privacy policy