JamesArgo

Mathematical talent is so useful, and so few have it. I hope genetic interventions will help in this next century.

Both the Schrodinger equation and inflationary cosmology predict alternate worlds filled with alternate yous. MWI refers to the former. The author of this piece is talking about the later.

If anyone is looking for a simple introduction to the Everett interpretation I'll point you to this short lecture by James Ladyman: https://www.youtube.com/watch?v=cgXVEeL9tzM

It wouldn’t be an exaggeration to say that a front line, if not the trench of the global gender war, is in Silicon Valley. In that sense, Silicon Valley culture echoes the Wolf of Wall Street culture in the ’80s and ’90s.

Signal the right politics and even this nonsense gets upvoted. Scott Alexander wrote about why differences in outcome in highly technical fields might not imply a vast economically-counterproductive misogynistic conspiracy here:http://slatestarcodex.com/2015/01/24/perceptions-of-required...

Reality is under no obligation to live up to our ideals.

For anyone interested, here's Scott Aaronson's response to the paper: http://www.scottaaronson.com/blog/?p=1767#comment-103591

OK everyone: At several people’s request, I’ve now taken a look at arXiv:1403.7686, and I can confirm that it’s complete garbage. The author is simply mistaken that solving the Schrödinger equation is “NP-complete” in any interesting sense: his argument for that seems to rely on a rediscovery of the adiabatic algorithm, but he doesn’t mention that the spectral gap could be exponentially small (and hence the annealing time could be exponentially large)—the central problem that’s been the bane of Farhi and his collaborators (and, of course, of D-Wave) for the past 15 years. Also, even if you thought (for totally mistaken reasons) that quantum mechanics let you solve NP-complete problems in polynomial time, that might (or might not) suggest to you that quantum mechanics should be replaced by something else. But until you’d actually found a replacement, and given some sort of evidence for its truth, I don’t see how you could claim to have thereby “solved the measurement problem”!! As additional problems, the author appears to conflate the P vs. NP problem with the question of whether NP-complete problems can be efficiently solved in the physical world, a common novice mistake. And also, he seems comically unaware of everything that’s been discovered in quantum computing theory over the past 20 years relevant to the issues he’s writing about—as if he just emerged from a cave.

Macroscopic decoherence has explained away Schrodingers Cat. It's been a non-mystery for years.

No one has every believed in the basilisk, save for maybe Yud. Everyone on LessWrong thought it was dumb from day one. LessWrong, now mostly dead, was a fun community full of smart people. That article is trying to make the place out as a cult, but no one bought any of that stuff, not even Roko. He was just playing around with local memes. It's pretty gross how this whole basilisk thing has turned out.

It's assumed a rational agent judges its actions by how likely said action are to align the world with their fundamental values. For this reason, it is argued, a rational agent will not change its terminal goals. For example, suppose we build an agent that desires only to make shoes. That is, its ideal universe would consist of all available matter being converted into sneakers. Would it realize this is stupid? No. It judges it actions and thoughts only by how many shoes they are likely to bring into being. Any thought against this would be judged, correctly, as a negative contribution to shoe production and so wouldn’t be implemented, and likely would remain unthought. The best paper on this by far it Basic AI Drives by Stephen M. OMOHUNDRO: http://selfawaresystems.files.wordpress.com/2008/01/ai_drive... I urge you to read it. It really is a wonderful paper.

The correct analogy would be a jail run by a bacterium.

Most chefs are male - people don't care because its a low status job. Programming is now high status, so now this gender gap is an issue. As many feminists have "misogynistic conspiracy" as their default explanation for any differnces in outcome among genders, they're not in a very good epistemic position to judge these things. Unjust discrimination is economically counterproductive, I don't expect ruthless corporations to support such a conspiracy when defection is more profitable. Aside from blinded applications and maybe interviews, not much needs to be done. We are building businesses here, not collecting data for a census.

If this means anything, it means we should experiment with newer forms of government, not be happy with the best of a bad lot.

To feed gwern's smug sense of superiority to the rest of humankind

Really?

And let us add, people create art because it's fun, which is reason enough for almost anything.

He's arguing against subsidising art, not for stopping all art production.

With head tracking it can look quite good. Though that only works for one person.

Scott S. Alexander wrote a great blog post about this called Society Is Fixed, Biology Is Mutable: http://slatestarcodex.com/2014/09/10/society-is-fixed-biolog...

We can't measure things we don't understand.

What? Was everyone blind before the discovery of quantum mechanics? If we can't measure things we don't understand, how the heck would we come to understand them? Through their affects on the world? "No, you say, "we can not perceive things whose causes we don't understand." Where did you get your epistemology? Was it The Secret?

So basically, we can't make a photodetector without first being able to create a sun?

Slightly-misleading TLDR: Iterated embryo selection is sort of test tube reproduction and selectiong. If you can create gametes from stem cells, then create many more embryos with these gametes and then score each embryo, repeating the process selecting for those that have the best genetic correlates of cognitive ability, you can perform the equivalent of a 2000 year eugenics experiment in a test tube in about six months. You could potentially get far, far higher IQs than 160.

Prerequisites for this are cheap-gene sequencing, very, very large databases of genomes tagged with their donor’s cognitive measures, and advances in human stem cell research.

This is just regression toward the mean.

Human intelligence is heritable.

This is indisputable. Denying IQ and its heritability is equivalent to denying global warming. It’s settled science. Dysgenics is certainly possible. I think technology like iterated embryo selection and genetic engineering will come into common use before dysgenics kicks in. China is already investigating these possibilities, and their culture does not have the same memetic immune response to such ideas. After one country makes it legal, competitive pressures will force the rest of the world to do it, too. Iterated embryo selection could conservatively raise IQ by 60 points, likely much much more: http://www.nickbostrom.com/papers/embryo.pdf

A government would have to be supernaturally incompetent to allow such a gap to develop.

You've got this normative halo around "intelligence". All intelligence is, is an agent's ability to satisfy its utility function.

But if it continues to work on the paper clips and decides that they are still needed even after people are gone - how is that super intelligent?

That is the correct action if all it cares about is making paperclips. Creating paperclips satisfies its utility function. Fretting over the existential issues of creating paperclips long after the last piece of stationary was destroyed would be judged of negative utility.

The AI is modelled as an agent with goals. If you want to argue that "intelligence" means "has a terminal goal exactly matching the values of humanity," then I suppose a paper-clip maximizer is not intelligent, but that's a non-standard definition of intelligence. The more common definitions for intelligence go something like this: The capacity of an agent to select and implement actions that bring about its goals. Here the "goals" could be anything. That is, any goal could be combined with any level of capability so long as the agent is intelligent enough to comprehend its goal.

If you have to specify the goals very carefully then the AI it's not super intelligent by definition.

When you're designing the agent, you have to select something like a goal, whether it's a fitness function or the maximization of a reward signal. Evolution certainly selected our goals. Love, self-expression, sex, fun, intellectual stimulation, the beauty of social interaction – these all reflect features of our ancestral environment. It’s a little naive to think there is a One True Morality that all agents stumble upon. It’s even sillier to think it happens to reflect our values

The intelligent part of the paperclip maximizer selects actions and weights them in the context of its utility function, to maximize paperclips. Any thought of changing its goals would be judged of negative utility, as that would lead to less paperclips.

I like this take. Last century we built many amazing bodies, this century we give them brains.

I never took Bitshares seriously. I thought their licensing strategy was insane so I dismissed them. Then I read their delegated proof of stake paper. It's one or the more rational, practicable alternatives to POW I've seen proposed.

He's an extraordinary accomplished writer. I recommend his short story collection, Axiomatic, as a great introduction to his work.

This is an example of useless guilt. Strive to blind yourself when there are consequences of your bias, otherwise acknowledge it. Beauty leaks into our perceptions of all other personal qualitative - both men and women are affected by it. Short of brain-modifcation, you will find it impossible to disregard beauty so don't feel guilty about it. And don't think "I'm glad I'm not one of those guys." Women are just as afflicted by "lookism" as Ted Chaing called it. Men and woman are people. People suck.

There are other biases like "force of personality bias", too, where we assume charisma is heavily correlated with correctness, which isn't necessarily true. In terms of solutions, things like blinded hiring and internal prediction markets might help mitigate these. However, if we are going to experience beauty and charm we have to accept the fact that these qualities will affect our opinions of those who possess them. There's no way around this without changing the nature of our perceptions of these phenomena, and thus the phenomena themselves.

Or even caring about people's emotions and sense of self, seems to have never occurred to any of them that I have met.

But that's not relevant to a community focusing on rational discussion. And it's the exact opposite of what cults do - they try to shower their followers with love. Not caring too much about people's emotional attachments to their beliefs is appropriate on LessWrong. It is inappropriate outside of that context.

I do not understand why there are so many haters of LessWrong. It's not a cult. No one on that site has ever bought into the Roko idea. Also, Roko's post is poorly explained there. However, everyone thought it was nonsense at the time and still does. Imagine if the most inane nonsense posted to hacker news was used to dismiss everyone who posts here.

As for nanotechnology. It is possible - biology is a proof of that - though some of the early visions may have been naive. It's likely that artificial systems can be created using a broader pallet than natural selection does, and that these technologies will be powerful.

That it is justifiable to kill a man if it would remove a spec of dust from the eye of every other person on Earth

No, he wrote that it was justifiable to torture a man to prevent 3^^^3 dust specs from floating into 3^^^3 eyes. 3^^^3 is a mind-bogglingly large number - far far far far far more than everyone on earth.

Use lead-free solder folks; I strongly believe we should keep lead solder the hell away from children.

Yes. I just noticed that - an idiotic oversight. I added that in. I think in general I'm more wrong than I thought I was. But at some point it breaks. If all the clones have an IQ of, say, 50 there's no way they would be able to survive. To take it to an absurd extreme, suppose we cloned chimps instead of people. The vast majority of the new 100x resources would end up in human hands.

That is, it seems obvious to me that humans would find a use for that land that is more efficient than selling its outputs to the chimps, whose value would be absurdly low - 100 chimps being worth less than one human. I think it's more likely the humans would up their land use by 100 (even supposing a static population) than attempt to sell food for nearly worthless chimp labour.

You're starting to sway me. But suppose we cloned the least-productive 0.001 percent of the population such that they outnumbered us all one hundred to one. Suppose, too, that all resources on earth were multiplied by the same amount. Do you think that would end well for the clones? Would Honda start making cheaper cars for the clones? What forces are at work that would guarantee the survival of these unproductive clones?

You're assuming there's not a more economically efficient use for farmland than feeding really, really, really poor people. The floor for the price will be the opportunity cost of using the land to grow food rather than, say, covering it with solar panels and exporting computation or whatever else can be dreamed up. At some point, it costs more to feed people than their economic worth over their lifetimes. I think in the long term, computers and robotics will be ridiculously efficient replacements for labour.

I don't really expect everyone to starve to death - and to be honest I think there are more-worrying implications of AI than this: http://selfawaresystems.files.wordpress.com/2008/01/ai_drive...

I do, however, want to point out that's there no economic law preventing the majority from being far worse off.