I'm considering freelancing in London - I'm also into machine learning and algorithms. I would love to hear about your experience - has it been easy to find work in this area?
I work for a media monitoring and analysis company. It's a special arrangement, and the research is potentially beneficial to them, so it's probably a one-off.
I just went from five days to three days a week in my job in industry in order to work as a researcher the other two days. I had to take a pay cut to do so, and I was lucky that my company was happy for me to do this.
So far, it's working out great - I do the research work from home, occasionally commuting to uni, and it's great to have the balance of academia and real life work.
This reminds me of when I was helping a friend with a maths problem. He would say, "Is it this?", and I would say "No", and he would keep guessing. I suppose most people just don't think to ask why.
I've collected what I think are the most important files. They compress very well because they seem to be very inefficiently stored (lots of repetition). The compressed collection is now on github.
I have the files but I haven't uploaded them as they are very big. I imagined one of the other programs would create the files, but I've yet to discover it. I believe these files are precomputations that are meant to speed up the algorithm.
To be honest, most of the time I didn't have a clue what he was talking about, and I was by far the most mathematical of the family! He wasn't a great teacher, but he was certainly an inspiration. And apart from all this, he was a great uncle, and great company.
There is a year or two of work on top of the documentation I have included. Also, I think it is likely that there are some incremental improvements to the existing algorithms that my uncle put in place without necessarily documenting. Whilst these might not be theoretically interesting, their combined effect could be significant. There may also be some new maths in there too, as he was undoubtedly a brilliant mathematician.
That's a good question, and I'm not entirely sure of the answer. I'm pretty sure my uncle believed it did, but I'm not sure how in touch he was with the state of the art, as he didn't use the internet. His idea of research was photocopying articles (and whole books, in particular I remember vol 2 of Knuth's TAOCP) from the British library.
My plan is to get some of these algorithms running and compare running times with the other open source ones available.
The point about a quantum leap is that it is discrete, rather than continuous, so, for example, a "quantum leap in understanding" implies a progress in understanding that cannot be broken down into a progression of small increments of understanding.
This is similar to what I was once told is the foundation of the Islamic sciences: logic, linguistics and rhetoric. These are essentially the "meta" sciences, which teach you how to think.
I think PhDs are an end in themselves, not a means to an end
I agree completely. I advise people to only do a PhD if they'll be doing what they love. It's the one chance you get to do exactly what you want, so make it good.
At worst, BitCoin at this point is a scam where people who got involved with it early who are sitting on lots of BitCoin need to keep up a level of interest to cash out BitCoin for coin, as the curious poke around with it.
Like any currency without intrinsic value (like paper money) it only has value when people believe it has value.
You're right - I talk about how to incorporate logical representations in my thesis, but the reviewers asked me to remove it as it wasn't complete enough. We had some more thoughts about the correct way to do this in this paper: http://homepages.feis.herts.ac.uk/~dc09aav/publications/iwcs...
This is a site that I made a while ago in python/mysql. I ported it to App Engine a few months ago, but due to a disagreement with my hosting provider, I wasn't able to transfer the domain name until today.
I hope to write something soon about my experience of porting to App Engine.