4 ms·

I think I would argue that this is beyond an AI-complete problem, it is an unanswerable problem. For instance, consider the following spectrum of ideas: - The

by darawk 10y ago

I think I would argue that this is beyond an AI-complete problem, it is an unanswerable problem. For instance, consider the following spectrum of ideas: - The earth is flat - 9/11 conspiracy theories - Barack Obama was born in Kenya - Anti-vax - Climate change denial - Anti-GMO - Denial of underlying quantum randomness in the universe - The Clinton Foundation has been or is involved in some shady dealings - Muslims want to implement Sharia law in Europe - The minimum wage decreases employment - Donald Trump has sexually assaulted a number of women - Monetary policy impacts the real economy - The sun will, on balance, probably rise tomorrow I've sorted this in a (rough, personal) order of increasing plausibility / conformance with facts. Where is the line between misinformation and alternative information? As a human, it's pretty unclear to me.

hawkice 10y ago
Man, it must be me, but you ranked "Monetary policy" WAY more likely than I would (rates near zero for so long doing so little must at least put SOME doubt in your model), and "denial of quantum randomness" way less credible than I see it as. Like, man, reasonable people can disagree about pilot waves and many worlds. Are there some confusing, emotionally laden, difficult questions of fact in there? Yes. Are some things ultimately unknowable (like what people want, which involves scrying into their minds)? Yes. These, as it turns out, are not the things I can't help but be side-tracked on. I'm diverted by you claiming monetary policy has an effect with nearly the certainty of the sun rising.
- darawk 10y ago
  Haha, sorry I knew this list would cause this sort of trouble. I'm a big believer in pilot wave theory myself, it's just very much not the 'mainstream' view. And I didn't mean to imply that the last two were close together, either. I do think monetary policy probably has some effect (whether we understand or can model that effect is another matter), but I didn't mean at all to imply that it was anywhere near the certainty of the sun rising, it just happened to be the closest :). I think perhaps we disagree more about the probability of these two: - The Clinton Foundation has been or is involved in some shady dealings - Muslims want to implement Sharia law in Europe Than the other two. I think both of those are, in some sense, almost certainly true. There are Muslims who want to implement Sharia law in Europe. The Clinton Foundation has been involved in transactions that at least have the appearance of moral uncertainty. The headlines that I listed are just overly strong statements/generalizations of those (IMO undeniable) facts.
  - hawkice 10y ago
    I keep getting sidetracked by other issues, but you're statements are about certainty and uncertainty and where do you draw the line for heavy-handed intervention, so I might as well get into them: "undeniable" does not mean what you think it means. Example: http://www.cbsnews.com/videos/hillary-clinton-defends-foundation http://www.cbsnews.com/videos/hillary-clinton-defends-founda... There, Clinton denies the foundation has been engaged in anything shady. Boom, definitely not undeniable #hackernewsdrama, hahaha. If you want a disagreement, though, I prefer many-worlds over pilot wave theories, largely because I'm not entirely sure _what_ the propagation speed of the pilot waves would be. 'Greater than the speed of light' either (1) doesn't really narrow it down at all or (2) narrows it down to having no options at all. But they should have _a_ speed, because I'm not sure you can even have standing waves if the disturbances propagate instantly.
- drumdance 10y ago
  Rates near zero is part of the reason the stock market has been on a tear the last few years. It also was a driver of the housing bubble. It didn't cause the hyperinflation that a lot of people thought would come, but the Great Depression is Bernanke's area of expertise, and he had sound theoretical reasons to believe QE wouldn't spark serious inflation.
  - dragonwriter 10y ago
    The near zero rates occurred in response to the economic crisis after the housing bubble burst, not a driver of the bubble.
    - hueving 10y ago
      Rates were also comparatively low during the 2000s to juice the economy after the tech bust and 9/11. They got even lower after the crash and QE, but they were the lowest they had been in the 00s since the early 60s.
cmdrfred 10y ago
I don't deny quantum randomness per say, I just deeply hope that it isn't true for some reason.
bogomipz 10y ago
Could you elaborate on what "AI-complete" means? I have not heard this term before. I am assuming it similar to NP-Complete but in the domain of AI? If this correct I assume that there are also problem that are classified as AI-hard?
- darawk 10y ago
  AI-complete, I believe, means 'full AI'. An AI-Complete problem is a problem that can't be solved without creating full human-level (or above) AI.
  - astrodust 10y ago
    That's a meaningless term since defining "human-level" is already a lost cause. Some people can multiply ten digit numbers in their head. Some can't even tell you what a number is. There's a very wide bell-curve here.

by darawk 10y ago

hawkice 10y ago

Man, it must be me, but you ranked "Monetary policy" WAY more likely than I would (rates near zero for so long doing so little must at least put SOME doubt in your model), and "denial of quantum randomness" way less credible than I see it as. Like, man, reasonable people can disagree about pilot waves and many worlds. Are there some confusing, emotionally laden, difficult questions of fact in there? Yes. Are some things ultimately unknowable (like what people want, which involves scrying into their minds)? Yes. These, as it turns out, are not the things I can't help but be side-tracked on. I'm diverted by you claiming monetary policy has an effect with nearly the certainty of the sun rising.

darawk 10y ago
Haha, sorry I knew this list would cause this sort of trouble. I'm a big believer in pilot wave theory myself, it's just very much not the 'mainstream' view. And I didn't mean to imply that the last two were close together, either. I do think monetary policy probably has some effect (whether we understand or can model that effect is another matter), but I didn't mean at all to imply that it was anywhere near the certainty of the sun rising, it just happened to be the closest :). I think perhaps we disagree more about the probability of these two: - The Clinton Foundation has been or is involved in some shady dealings - Muslims want to implement Sharia law in Europe Than the other two. I think both of those are, in some sense, almost certainly true. There are Muslims who want to implement Sharia law in Europe. The Clinton Foundation has been involved in transactions that at least have the appearance of moral uncertainty. The headlines that I listed are just overly strong statements/generalizations of those (IMO undeniable) facts.
- hawkice 10y ago
  I keep getting sidetracked by other issues, but you're statements are about certainty and uncertainty and where do you draw the line for heavy-handed intervention, so I might as well get into them: "undeniable" does not mean what you think it means. Example: http://www.cbsnews.com/videos/hillary-clinton-defends-foundation http://www.cbsnews.com/videos/hillary-clinton-defends-founda... There, Clinton denies the foundation has been engaged in anything shady. Boom, definitely not undeniable #hackernewsdrama, hahaha. If you want a disagreement, though, I prefer many-worlds over pilot wave theories, largely because I'm not entirely sure _what_ the propagation speed of the pilot waves would be. 'Greater than the speed of light' either (1) doesn't really narrow it down at all or (2) narrows it down to having no options at all. But they should have _a_ speed, because I'm not sure you can even have standing waves if the disturbances propagate instantly.
drumdance 10y ago
Rates near zero is part of the reason the stock market has been on a tear the last few years. It also was a driver of the housing bubble. It didn't cause the hyperinflation that a lot of people thought would come, but the Great Depression is Bernanke's area of expertise, and he had sound theoretical reasons to believe QE wouldn't spark serious inflation.
- dragonwriter 10y ago
  The near zero rates occurred in response to the economic crisis after the housing bubble burst, not a driver of the bubble.
  - hueving 10y ago
    Rates were also comparatively low during the 2000s to juice the economy after the tech bust and 9/11. They got even lower after the crash and QE, but they were the lowest they had been in the 00s since the early 60s.

cmdrfred 10y ago

I don't deny quantum randomness per say, I just deeply hope that it isn't true for some reason.

bogomipz 10y ago

Could you elaborate on what "AI-complete" means? I have not heard this term before. I am assuming it similar to NP-Complete but in the domain of AI? If this correct I assume that there are also problem that are classified as AI-hard?

darawk 10y ago
AI-complete, I believe, means 'full AI'. An AI-Complete problem is a problem that can't be solved without creating full human-level (or above) AI.
- astrodust 10y ago
  That's a meaningless term since defining "human-level" is already a lost cause. Some people can multiply ten digit numbers in their head. Some can't even tell you what a number is. There's a very wide bell-curve here.