Tag Archive: artificial intelligence

The exam is over, and I think it went okay, but I’m not at all confident that I’ll have passed this one – hopefully the results will be released around the start of December like they were last year, I’ll be very interested to find out whether I’ve managed to scrape a Pass 4! Anyway, now that it’s over and done with, it’s time for a good navelling-gazing session about M366 as a whole…
Continue reading


I finally got TMA04 for M366 back this morning, and was very happy to find that I’ve got my best mark of the entire course for this one – 93%! Which is a well-timed confidence boost, since my M366 revision has been challenging to say the least! I’ll be happy with a Pass 4 in the exam, to be honest. Let’s just hope there are plenty of questions about Block 5, since that seems to be my strong point…

I got the MS221 TMA04 back a while ago, but I only had a quick look at the score and the summary comments, as I didn’t want it to distract me too much from the M366 exam prep I’m doing at the moment. I got a disappointing 88% for this one, which particularly worries me since Block D was my favourite part of the course, and I expected to do a lot better than that. I can’t believe I scored better on the calculus TMA than this one!

Well, I’ll have plenty of chance to go through the detailed feedback and find out exactly what went wrong with my understanding of Block D, once the M366 exam is over and done with next Wednesday. Until then, it’s back to neural networks, learning rules and search algorithms for me!

I can’t believe it’s only 3 weeks until the M366 exam! This course has really dragged at times, but now that I’m into the revision period the days are just flying by! This week I’ve been making notes on the learning outcomes for each block, and trying to condense my notes as much as possible. There’s something very satisfying about squashing a nine-month course into a nine-page Word document!

I’ve been doing some MS221 revision this weekend too – I decided to have a go at Part 1 of the 2007 past paper, and I was interested to see that the questions don’t completely conform to the pattern in the 2008 paper. I was particularly blind-sided by the appearance of a question on combinations/permutations and binomial coefficients, since I’d forgotten they were actually covered in Chapter B1.

With MS221, I’ve mostly been concentrating on getting faster at answering questions – there are 12 questions in Part 1 of the exam, and I’m aiming to spend 2 hours on that bit, so I need to be able to do each short question in about 10 mins. So far, I’m fast enough when it comes to my favourite topics (Block D, I love you!), but in my weakest areas I have to spend far too long flipping through the Handbook and scratching my head.

In an attempt to speed the process up, I’m putting together a kind of condensed version of my most frequently-used bits of the Handbook, so that I can transcribe it into the blank front page of the actual book and take it into the exam. It’s going to be quite a challenge to hand-write all these notes into the Handbook, but hopefully it’ll allow me to spend more time puzzling over the actual questions and less time flicking through the book. And it’ll give me an opportunity to use my many colourful gel-pens – my Handbook will probably be the most gaudy book in the exam hall!

I got 85% for this one, so it’s a distinction but only just! By my calculations, I need to get at least 82% for TMA04 in order to get a distinction for the Overall Continuous Assessment Score – but to be honest, I’m probably going to end up with a Pass 2 in the exam, so TMA04 might not even matter that much anyway!

I’m starting to feel like I’m just not good enough to be doing M366; I lost a lot of marks on Question 5, the case study question, so it seems like I just don’t really understand the material well enough to apply it to practical scenarios. I know I’ve said it before, but I really wish I’d taken M359: Relational databases: theory and practice instead – I’m sure that would have presented its own difficulties, but in hindsight I think it would have suited my interests and abilities a lot more than M366!

M366 is a very strange course. I haven’t really enjoyed reading the course texts themselves, but the assignments are much more interesting. I spent ages fiddling around with the rabbits simulation for TMA02, and I’m finding the neural network experiments in TMA03 even more rewarding. Yes, I’m one of those weirdos who enjoys collecting and analysing data – so compiling the big network performance tables and assessing the accuracy of the perceptrons in Questions 1 & 3 was brilliant fun for me.

At the moment I’m working through Question 4, the one about self-organising maps, but unfortunately I’ve not been able to get my SOM to work properly so far – I’ve no idea why, but JavaNNS just doesn’t respond when I click the “Learn All” button to train the network. I’m probably missing something obvious, so I’ve emailed my tutor to see if he can figure out what’s going wrong – and in the meantime I guess I’m taking a bit of an enforced break. Can’t wait to get back to it, though!

It’s such a weird feeling – in the space of a week, I’ve gone from complaining that M366 feels like a second job (while I was reading through Block 4), to being so eager to get back to my TMA03 experiments that I find myself daydreaming about it during my actual job! I hope this means that I’ll enjoy the exam just as much…

UPDATE: Victory! I’ve finally managed to get my SOM for Question 4 working – reinstalling JavaNNS seems to have done the trick (but I still have no idea what caused the problem in the first place). Now, on to Question 5!

You’ll notice there isn’t the characteristic exclamation mark on the end of this post’s title – I’m a bit embarrassed of these two assignments, since both got grades quite a bit lower than the previous ones. For M366 I got 84% (just 1% short of a distinction, dammit!), and for MS221 it was 89%. I’m actually more ashamed of the MS221 score, because almost all of the lost marks were due to stupid errors, which I absolutely should have picked up at the proofreading stage. My tutor’s comments were very nice, but I think he must be getting a bit frustrated with the trivial errors I keep making. Basically, his general feedback was:

  • Factorise expressions where possible
  • Check your work (Mathcad is useful for this)
  • Read the question carefully!

All very good advice, which I have been striving to apply today while working on TMA03. I’ve been checking everything, every little calculation, either in Mathcad or Wolfram Alpha – so far, I’ve caught some really stupid slip-ups, so hopefully I’ll be able to manage more than 89% for this assignment!

The M366 assignment got a lower mark, but I’m happier with that one because the lost marks were due to a genuine lack of understanding about various little details of the subject matter, rather than plain old carelessness. Actually, I get the impression that most of my lost marks were caused by me not fully understanding the questions themselves; some of them wanted slightly different or fuller answers than I expected. For example, in the question about the NetLogo rabbits simulation, I lost three out of six marks because I didn’t mention the details of the code behind the features I was talking about – I described the simulation at too high or abstract a level, without backing up what I was saying with a description of the “nuts & bolts” side of things. Which is an odd mistake for a programming student to make! Still, hopefully by the time the exam comes around I’ll be much better at interpreting the questions correctly. And at least there’s no NetLogo coding to mess about with in the exam!

I’ve been anxiously waiting for this assignment to be marked for a couple of weeks now, and I think it’s actually the most worried I’ve ever been about a TMA. But thankfully, I got my assignment back today, so the nervous waiting is finally over! I got 91% for this one, which looks a bit rubbish compared to the scores I’ve had on previous computing courses, but then again, it is a level 3 course, so of course it should be more difficult than M263 or M255, etc.

Continue reading

Dr Perceptron from Futurama (picture from theinfosphere.org)

Dr Perceptron from Futurama (picture from theinfosphere.org)

Not Dr Perceptron, that is – the perceptrons I’ve been grappling with this week are the ones featured in Block 4 Unit 2 of M366, which I think has been pretty much the toughest part of the course so far. Not boring, just very dense and packed with detail. This was the first unit that still had me scratching my head at the end of my first read-through; it wasn’t until I’d gone through the unit again to make sure I’d covered all the Learning Outcomes that I actually started to feel comfortable with the material.

And one thing that particularly had me scratching my head was a curious glitch in the PDF copy of Block 4, which somehow omitted quite an important element of Figure 2.44, a section of which is pictured below. Can you spot what’s missing?

Continue reading

I’ve taken a break from studying this week due to a nasty cold which turned me into more of a Mucus Machine than a TMA Machine, but today I’ve been getting back to M366, and in particular getting stuck into Block 4: Neural Networks.

Unit 1 Section 3 is about recognising and classifying patterns, and touches on the human ability to recognise noisy or incomplete patterns. The ability to spot patterns has always impressed me (although it does lead to the bizarre excesses of religious-themed pareidolia), and I was especially interested to read that humans are apparently pretty good at recognising noisy patterns aurally as well as visually.

Humans are great at seeing patterns everywhere. The downside is seeing the Face on Mars or the Virgin Mary in your cheese on toast, but the upside is spotting building-faces like this one.

Humans are great at seeing patterns everywhere. The downside is seeing the "Face on Mars" or the Virgin Mary in your cheese on toast, but the upside is spotting funny building-faces like this one.

Section 3 briefly mentions some experiments done by Richard Warren in 1970, in which test subjects were played a recording of a polysyllabic word, with one syllable obscured by a loud click. Although the test subjects reported that they had heard a click, they could not correctly identify which syllable had been obscured; even odder, they reported hearing the whole word, including the missing syllable. Fascinating!
Continue reading

<a href="http://en.wikipedia.org/wiki/Shakey_the_Robot">Shakey the Robot</a>, star of M366 Block 2. He might be a bit clunky, but I still think he's great.

Shakey the Robot, star of M366 Block 2. He might be a bit clunky, but I still think he's great.

Today I finally worked up the nerve to submit M366 TMA01, after a good few weeks of tinkering and fiddling with it. I think I’m more nervous than usual about it because M366 is a very different kind of course to the ones I’ve done in the past, so I’ve no idea how badly or well I’ll do. I suspect I’ll get somewhere between 60%-80% for this assignment, but I wouldn’t really be surprised if it turned out to be a low Pass 3; I’m not confident at all about my ability to provide the kind of answers the TMA seemed to be asking for, so I’ll be relieved with anything above 40%!

Speaking of M366, my tutor replied to my email about TMA02 this morning, so I should be able to get that wrapped up tomorrow. It looks like I might have gotten the wrong end of the stick about Q4 (a)(ii), the one that requires 400 words when I’ve managed about 80 – the Curse of the Vaguely-Worded Questions seems to have struck M366 again. I’m not sure how much I can actually divulge about the disparity between what I thought the question was asking for and what my tutor has advised me to write about – but in any case, if anyone is finding it hard to write 400 words for this question, I would definitely recommend asking your tutor for guidance, in case you’ve interpreted the question the same way I had.

Continue reading