Yes, I think that was better, because the ground truth is Kepler's third law and jimrandomh pointed out your method actually recaptures a (badly obfuscated and possibly overfit) variant of it.

"Dimensionality" is totally relevant in any approach to supervised learning. But it matters even without considering the bias/variance trade-off, etc.

Imagine that you have an high-dimensional predictor, of which one dimension completely determines the outcome and the rest are noise. Your shortest possible generating algorithm is going to have to pick out the relevant dimension. So as the dimensionality of the predictor increases, the algorithm length will necessarily increase, just for information-theoretic reasons.

Using a high-powered black-box technique to regress a one-dimensional continuous outcome against a one-dimensional continuous predictor seems misguided.

If you want to characterize how well your evolutionary learning idea works, try it on data that you've generated, where you know the "underlying math". See if you can recover the program that generated the data or one that's equivalent to it. Or try it on really big, messy data where no one knows the right answer and see if you/it can do better than the obvious competitors like SVM, k-NN, CART, etc.

The middle ground of working on an easy/messy problem, where any sane method will give you and adequate answer but there's no known ground truth, is not going to make a very compelling story.

I had an 80$ logitech keyboard (the illuminating short-stroke like a notebook variety), and when it began to deteriorate I swapped it with a 10$ Walmart special that was a slightly curved Microsoft one. I had been playing around on this typing speed site and was surprised to find that on the 5th attempt I had beaten my previous record with this new keyboard.

If I had a variety of keyboards at my disposable I think it would be an interesting exercise to test them in this fashion.

What is your position on Will Newsome?

I frequently find Will's contributions obscurantist.

In general, I find obscurantism at best tedious, and more often actively upsetting, so I mostly ignore it when I encounter it. Occasionally I engage with it, in a spirit of personal social training.

That said, I accept that one reader's obscurantism is another reader's appropriate level of indirection. If it's valuable to other people, great... the cost to me is low.

At this point the complaining about it by various frustrated people has cost me more than the behavior itself, by about an order of magnitude.

If your goal is to lower your credibility, why do that in the context of talking about credibility?

Rank Theory of Depression

I believe I have a few results of this nature in my 23andme profile but, like most results there, they indicate e.g. that I might gain an extra .5 pounds compared to average on a high fat diet.

I got a kick when I logged in there and it said something to the effect of 'see how your genes affect your weight' and after entering height, age, and weight it told me that my genes were responsible for 2lb (whatever that meant).

It does also note lactose tolerance, alcohol and caffeine enzymes, coeliac disease risk, etc.

No, I think the idea is to do coarse-grained scans, which the superintelligence will have to heavily process in order to infer the original brain structure. (Yeah, it's not clear this is possible even with a whole universe worth of computing power and whatever algorithmic breakthroughs a superintelligence might come up with.)

Ok so I'm presuming that an extremely fine grained scan stored with some naive compression is massively more than 10^14 synapse-bits. In order to store all that now in the information theoretic minimum, don't we need some kind of incredibly awesome compression algorithm NOW that we simply don't have?

EY changed it in the published version to:

"The AI neither hates you, nor loves you, but you are made out of atoms that it can use for something else."