Evil genius

Messed up weather

New Zealand is having a record cool summer, while Switzerland is having a record warm winter. For example, the forecast high tomorrow in Wellington is 16 C — the same as here in Lugano! For New Zealand in the middle of summer to be the same temperature as Switzerland in the middle winter is so screwed up I can hardly believe it's happening.
Evil genius

What Bill Gates wants for Christmas

I just read this, sort of an interview with Bill Gates. At one point he was asked what he wants for Christmas. His answer was lectures from here. Me too.

I depart for New Zealand in a few days. Man I hate that flight. Anyway, it will be nice being back in NZ again. I'm also spending almost two weeks in Rarotonga (tropical island) before heading to Los Angeles for a couple of days, and then back to Switzerland. Once I arrive back I'll have just 4.5 months before my PhD is over.

I'm starting to work out what to do next. Of course what I'd really like to spend the next 5 years doing would be creating a radical new artificial intelligence system. After working in two companies trying to do that, and having spent the last 4 years learning more about AI for my PhD, I know how I'd go about doing it and I'm reasonably confident that I could succeed. However I have no idea how I could finance it. I estimate that I'd need $1.5 m to get to the point of proof of concept. It would require about 3 people, 10 computers and 5 years work. Once proof of concept had been achieved money would be no problem — the company that is first to develop real AI will easily have a market cap north of a trillion dollars. Alas I don't know any angel investors with that kind of money and outlandish ambition.

Plan B. Currently this plan looks like doing a post doc, perhaps a part time one, in finance. I'd be based in Switzerland and would apply machine learning methods to problems in finance research.
Evil genius

IT crisis, depending on your perspective..

According to this article, in the UK,

"... in the past four years demand for IT and computer graduates has doubled while at the same time the number of students studying the subject has declined by a third."

I've also heard similar things from the US in recent weeks. Even India is supposed to be having a skills crisis according to a few articles I've read recently. Demand there is so out of match with supply that the annual wage inflation for IT workers is running at almost 20%. This is all good news for IT workers incomes. More evidence that shifting from academia to industry now might be a good idea.

If all this industry demand is for real, I find it strange that the demand for people doing theoretical AI research appears to be so bleak.
Evil genius

Busy, busy...

7 months left on the PhD. Lots to do and not much time to blog. I almost have a committee sorted out, and hopefully in a few weeks I'll move to "PhD candidate" status. I know that seems a bit strange at this point, but it's because I've joined the program after a few years at a research institute so I'm doing all this in a compressed amount of time. Anyway, over the next 6 months I need to get some more research done and of course write my thesis. Hopefully that won't be too much of a problem as most of the material will be ripped from my papers and then massaged to fit together into one document. Still it's a lot to do in the next half year — so don't expect to hear too much from me! In the middle of all this I'm also going on a trip to New Zealand for Christmas, then Rarotonga for a couple of weeks, and then Los Angeles for two days on the way home. Extra busy.

Another thing that I need to do is to work out what I'm doing next. One thing that got me thinking recently is the fact that a couple of very good AI theory guys I know who have done a few post-docs aren't having an easy time finding permanent positions. One has now taken another post-doc position as he didn't get any of the permanent positions, and another has given up completely and is now trying to find a job in industry. These guys are really good, and so seeing this is a bit of a worry. Given that I've worked for 5 years in industry, by the time I will have completed 2 post doc positions I'll be almost 40. If I then can't get a permanent position in a university and have to switch to industry, I'll be in trouble.

If I switch next year to business I'll be 33, and with 5 years industry experience already on my CV, I'm not looking too much like a career academic. That should make the switch a lot easier. Given that academia seems to offer little job security, little money and the pressure to publish all the time means that I'm unsure whether I'll ever be able to follow the longer term more ambitious things I'd like to work on, I'm seriously thinking about going to industry. Another problem is that in academia as a PhD grad you don't have much control over where you live next. You just have to follow the jobs shifting to a new city and possibly country every few years. That's bad news for relationships. I understand now why many academics' wives don't have much in the way of professional careers — in the prime of their career building years they are having to shift from one random town to the next.

Another thing is that I find academia to be quite limiting in some ways. If you propose something really radical with a high chance of failure, it's almost sure to be rejected by the funding agencies. You simply can't go away and try something crazy for a few years, instead you have to produce small units of publishable work all the time as this is a key measure of how your performance is judged. Of course if you are the head of the department you can do some crazy stuff as you have a permanent position. But by then it's probably too late as your really creative and crazy days are already over — you've been molded to the norm and blinkered by the dogma of your field. If I had $1 million in the bank, I'd shift somewhere cheap to live and embark on research so unconventional it would make respected academics wince.

Maybe somebody like Google would be interested in giving me a job that wasn't too bad. Or maybe I could find a niche and start my own business doing something interesting. In any case, I haven't lost my passion and belief that very powerful AI is not only possible, but that it's possible in the next few decades. That's the dream, now how do I live it? Getting a PhD in AI is just one step.
Evil genius

Agnostic to Atheist

For a long time I have called myself agnostic.

While I am almost certain that an Abrahamic god does not exist, I've been more reluctant to totally rule out more abstract notions of "god". Thus I didn't feel that my certainty in the non-existence of some kind of a god was high enough to be called an atheist. At a more practical level, where I grew up religious beliefs were pretty mild. You could be an atheist, or a theist, and nobody much would care. Even our prime minister is agnostic and nobody but a tiny ultra religious minority seems to care. As such, there was not much pressure to be explicitly against theism.

However I've decided to change my position. When most people say "God", what they mean is an Abrahamic style god, i.e. Jewish, Christian or Muslim. With respect to this concept of god I've already been atheist for some time. The second reason is that, unlike the mild religious atmosphere of New Zealand that I grew up with, on the world stage today conservative religious voices seem to be getting stronger and are influencing major world events. I think that these voices are moving the world in the wrong direction: away from reason, evidence and science, and towards a world where something is true just because that's what you have decided to believe. In such a climate I think that a clearer position is called for.

I am now an atheist.
Evil genius

Perfect pasta

I think I can now basically make perfect pasta sauce. The down side to this is that now when I go out to restaurants, unless they are good Italian places in Italy, I tend to wonder why I'm paying so much money to get something that's not as good as I make at home. Anyway, here's one pasta sauce recipe, more or less. See I can't tell you the exact recipe because I never actually measure anything, but I can give you the general idea...


Puree tomatoes. Forget that pre-made tomato sauce with stuff in, start from first principles.

Olive oil.

Honey. You need something to sweeten it a bit and I find that a honey with a mild flavour works well.

Onions, if you want. Amazingly, you can make a real nice sauce with just the three things above.

A tin of tuna works well with this.

EDIT: I forgot one thing, lemon juice. Add some lemon juice. Trust me.


Get a big pot with high sides. Stick the oil in. If you're doing onions, chop them up and fry them until they start to turn a little transparent. Stick the tomato puree in and add the lemon juice. Now bring it to the boil. It will start to go "plop-plop" and little bits of the puree with be leaping out of the pot and onto the walls. Good. Keep it boiling like that, but don't burn it or anything too drastic. Stir it from time to time. At some point add some honey so that it's very slightly sweet. If you taste it at this point it will still taste a bit bitter from the tomatoes, that fine. After a while the puree will reduce (and probably make your walls dirty) and stop going "plop-plop" and start making quick little bubbles without leaping out of the pot. Good. This is after something like 15 minutes I'd guess.

At this point you'll need to stir it a bit from time to time and it will start to get thicker and start to get slightly darker too. Make sure that there is enough olive oil so that the sauce is just very slightly oily. Keep on frying it. It's the frying that gives it a really good flavour as it makes the tomato caramelise slightly and gives it a sweet and very slightly smoky taste. The key thing now is knowing when to stop. You want to stop when it's thick enough to be pushed into one side of the pot and never flow back. Maybe even a bit more than that. Now take it off the heat. At this point I'd recommend adding a tin of tuna. Just drop the tuna into the hot sauce and mush it around a bit. Done.

I eat it with spaghetti. Italians claim that every different kind of pasta holds the sauce a different way and this all really important. I think they're delusional.

It might sound like a lot of work, but really it's not. Most of the time you can just leave it to reduce away while it's going plop-plop and making a mess all over your kitchen. If you get it right you should get a slightly oily sauce that's fairly thick, with a deep rich slightly sweet taste and a hint of bitterness. I eat exactly the above dish (sans onions because I didn't have any) about half an hour ago and can still taste the rich flavour in the back of my mouth.

I think the combination of the olive oil, plus some sweetener and frying it are what makes the magic happen. I watched part II of The Godfather movie the other week and sure enough, there was a scene about halfway through the movie where somebody cooking pasta sauce explains this very technique.