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 4 articles in this topic
Showing 4 of 175 articles.
Currently 2 other people reading this forum.


 UWA week 34 (2nd semester, week 5) ↓
SVG not supported

Login to reply
11:05pm Tue 23rd Aug, Tiziano W.

Hello, My partner and I were reading through the project brief, and we were just wondering if you would be able to clarify the first step of the initialization inputs. It states that the input will give n and p, the number of green agents and the probability of connections. We assumed these n, p values are denoting a binomial distribution, but we are unsure how we would setup the network using these. The brief states that every node has to have a probability of interaction with another node, but if we were to use the n p values as a binomial distribution, we would only be able to create a graph showing nodes as either connected or not connected, as a binomial distribution is boolean. We were wondering if you would be able to clarify how one would use the inputs to create a "weighted" network graph rather than simply a graph showing absolute connections between nodes. Thank you in advance for your time.


SVG not supported

Login to reply
4:16pm Wed 24th Aug, Ryan B.

You've stumbled upon a namespace conflict in CS (who would have thought??) The variables `n`, `p` are standard terminology in graph theory when we talk about constructing Random Graphs (https://en.wikipedia.org/wiki/Random_graph). The `n` is referring to the number of vertices (or 'nodes`, take your pick), and `p` is the probability that an edge between 2 nodes occurs. There are lots of ways to generate random graphs, that lead to different results. I'd look into the Erdos-Renyi model to start with as a simple approach. Another piece of advice is there are often graph libraries in that will do the random generation for you... Hope this helps!


SVG not supported

Login to reply
10:40am Fri 26th Aug, Mehwish N.

That is a very good reply Ryan! Those of you who would like to jump to implementing your agents first and then solving the initialisation stuff, can use the files that I have uploaded on LMS>Project2022 (also available on teams). Erdos-Renyi is indeed the easiest approach to generate graphs. If you want to implement more sophisticated solutions you can use other models (e.g., preferential attachment, planted partition or ERMGs). Please note you will not be penalised for not using a sophisticated network model, however those who implement sophisticated network models may end up in writing better test cases and producing close to real life solutions. A basic G(n,p), where n= number of nodes and p= probability of connections is fine. You may want to explore networkX in python or igraph (if you are working in python). I am only throwing pointers, at the end of the day it is your choice how you implement!


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

Login to reply
2:42pm Fri 9th Sep, ANONYMOUS

Hi Dr.Mehwish, are we allowed to use libraries to generate random graphs? Also when you said 'however those who implement sophisticated network models may end up in writing better test cases and producing close to real life solutions. ' do you mean that those who implemented sophisticated network models will get better marks than those who didnt?

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