与Eliezer Yudkowsky谈话的谈话成绩单。
面试官:我很乐意在你的一个“友好AI中的开放问题”中获得澄清。Benja Fallenstein的逻辑不确定性问题t对于代理商没有足够的计算能力推断出来的逻辑事实,必须对具有不确定性的不确定性。但是:我听到了一些不同的东西称为“逻辑不确定性问题”。One of them is the “neutrino problem,” that if you’re a Bayesian you shouldn’t be 100% certain that 2 + 2 = 4. Because neutrinos might be screwing with your neurons at the wrong moment, and screw up your beliefs.
Eliezer: 也可以看看如何让我说服2 + 2 = 3。
面试官: 确切地。即使在像一个这样的概率系统中金宝博官方贝叶斯网,它的组成部分是演绎的,例如,某些部分必须总和到一个概率,并且存在贝叶斯网结构中的其他逻辑假设,并且AI可能希望对那些的不确定性具有不确定性。这就是我称之为“中微子问题”。我不知道你认为的问题是多少,以及你通常如何谈论当你谈论“逻辑不确定性问题”时如何谈论。
Eliezer: I think there’s two issues. One issue comes up when you’re running programs on noisy processors, and it seems like it should be fairly straightforward for human programmers to run with sufficient redundancy, and do sufficient checks to drive an error probability down to almost zero. But that decreases efficiency a lot compared to the kind of programs you could probably write if you were willing to accept probabilistic outcomes when reasoning about their expected utility.
然后,这是一个友好的AI友好行动标准和自我修改的标准的大问题,在你将错误概率下降到零点后,我目前的所有想法都在证明事物的情况下仍然是措辞。But that’s probably not a good long-term solution, because in the long run you’d want some criterion of action, to let the AI copy itself onto not-absolutely-perfect hardware, or hardware that isn’t being run at a redundancy level where we’re trying to drive error probabilities down to 2-64或者某事 - 真的接近0。
面试官:这似乎可能与您经常谈论的事情不同,当您使用短语“逻辑不确定性问题”时。那正确吗?
Eliezer:当我说“逻辑不确定性”我经常谈论的时候更像是,你相信Peano算术,现在分配了Gödel对Peano算术的陈述的概率。或者您尚未检查过它,239,427是素数的概率是多少?
面试官:你在两个问题之间看到了大部分关系吗?
Eliezer: 还没有。第二个问题是相当基础的:我们如何近似逻辑事实,我们没有逻辑上无能为力?特别是当您对您正在运行的复杂算法具有不确定的逻辑信念时,您正在计算相对于这些复杂算法的自我修改的预期效用。
What you called the neutrino problem would arise even if we were dealing with physical uncertainty. It comes from errors in the computer chip. It arises even in the presence of logical omniscience when you’re building a copy of yourself in a physical computer chip that can make errors. So, the second problem seems a lot less ineffable. It might be that they end up being the same problem, but that’s not obvious from what I can see.