Well, I’m just back from a week spent in Launceston partaking in a Winter Unit in Advanced Algorithms and implementation thereof; so I expect this post to mostly talk about that.

Arriving at Launceston, we had dinner, and proceeded to an extended game of Bartog to kick things off, and introduce people to people that hadn’t already met other people. We stayed at the Australian Maritime College’s hall of residence, and in general, I rated the room as better than that of most other colleges I’ve stayed at, apart from the niggling annoyance that the heaters switched themselves off after two hours (permanently), and the blankets were far to thin to compensate for this. Luckily, the point of the trip was not to sleep.

There’s something particularly comforting about spending 15-16 hour days in computer labs at foreign campuses (NCSS, I mean you!), this was no exception. On most days, we’d arrive at the School of Computing building at 9AM, and not properly leave until 10-11PM; barring lunch and dinner breaks. This allowed for lots of problem-solving activities.

Monday consisted of a “Getting to Know C++” session, whereby seasoned C++ coders would get massively bored, and newbies would get overwhelmed. Luckily, I sat somewhere in the middle, and so learned quite a few useful hints. No specific algorithms were studied on the day, but rather, we looked at many ad-hoc examples of algorithm design. Most people seemed to grasp the concept of “Greedy” algorithm design, I certainly did.

Tuesday and Wednesday consisted of Graph Algorithms, which is a very comprehensive field at the best of times; we focused firstly on some extensions on things covered in 2nd-year; followed by an introduction to some NP-complete problems, namely, modification of a graph that contains no Euclid Circuit such that it does, and the Travelling Salesman Problem (whilst using Dynamic Programming). We also touched on things like Vertex Cover and Automaton Matching. All very complex, but I’m definitely looking forward to solving some of these.

Thursday consisted of a basic introduction to Computational Geometry, introducing things such as the Convex Hull, and distance between two pairs. I realised far too late in the piece that remembering some vector calculation concepts touched on in KMA152 would have been useful.

Today introduced some (trivial) arithmetic problems that involve writing classes that represent numbers in a non-native method (i.e. Big numbers, or rational numbers, for example). Would have been interesting, except that the concepts covered were no more complex than they were when I first experienced them (in ACPC, Year 10, incidentally).

Well, bedtime calls, so that’s clearly all the info you’re going to get from me today. I may post some discussions of problems posed as the week progresses — it should help with exam preparation for me :).