moridinamael(Matt Freeman)
The Parable of the King and the Random Process
Decision Theory with the Magic Parts Highlighted
Carrying the Torch: A Response to Anna Salamon by the Guild of the Rose
Complex Behavior from Simple (Sub)Agents
A Dialogue on Rationalist Activism
A Retrospective Look at the Guild of Servants: Alpha Phase
It has helped me to realize that the policies I implement to govern my behavior are implicit hypotheses about my behavior. The general form is “I predict that installing policy X will produce behavior Y.” In the past when my imposed self-regulatory policies have failed I have viewed it as a personal failure, a failure of the will, my failure to adhere to the policy. It helped a lot to realize that my hypothesis was merely wrong. I became free to update. I started learning from my failures.
It’s funny, I am totally sympathetic to everything you wrote here, yet all I can think is, “my daily life is chock full of people incapable of grappling with trolley problems or discussing torture concretely, why are you trying to make LessWrong more like real life?”
A Sarno-Hanson Synthesis
After seeing a number of rather gloomy posts on the site in the last few days, I feel a need to point out that problems that we don’t currently know how to solve always look impossible. A smart guy once pointed out how silly it was the Lord Kelvin claimed “The influence of animal or vegetable life on matter is infinitely beyond the range of any scientific inquiry hitherto entered on.” Kelvin just didn’t know how to do it. That’s fine. Deciding it’s a Hard Problem just sort of throws up mental blocks to finding potential obvious solutions.
Maybe alignment will seem really easy in retrospect. Maybe it’s the sort of thing that requires only two small insights that we don’t currently have. Maybe we already have all the insights we need and somebody just needs to connect them together in a non-obvious way. Maybe somebody has already had the key idea, and just thought to themselves, no, it can’t be that simple! (I actually sort of viscerally suspect that the lynchpin of alignment will turn out to be something really dumb and easy that we’ve simply overlooked, and not something like Special Relativity.) Everything seems hard in advance, and we’ve spent far more effort as a civilization studying asphalt than we have alignment. We’ve tried almost nothing so far.
In the same way that we have an existence-proof of AGI (humans existing) we also have a highly suggestive example of something that looks a lot like alignment (humans existing and often choosing not to do heroin), except probably not robust to infinite capability increase, blah blah.
The “probabilistic mainline path” always looks really grim when success depends on innovations and inventions you don’t currently know how to do. Nobody knows what probability to put on obtaining such innovations in advance. If you asked me ten years ago I would have put the odds of SpaceX Starship existing at like 2%, probably even after thinking really hard about it.
I have found that the more I use my simulation of HPMOR!Quirrell for advice, the harder it is to shut him up. As with any mental discipline, thinking in particular modes wears thought-grooves into your brain’s hardware, and before you know it you’ve performed an irreversible self-modification. Consequently, I would definitely recommend that anybody attempting to supplant their own personality (for lack of a better phrasing) with a model of some idealized reasoner try to make sure that the idealized reasoner shares your values as thoroughly as possible.
Took it. Nothing like a census/survey to make you feel like part of a community.
FYI
Advisory: RANDOM.ORG will be temporarily unavailable on Sunday 2012-11-04 due to a system upgrade.
Also, in retrospect, I misremembered my own age. It’s been … a busy year.
Flowsheet Logic and Notecard Logic
Proposal: Consider not using distance-direction-dimension words in abstract discussions
Spamming Micro-Intentions to Generate Willpower
It seems so obvious to me that the benefits of preschool would wear off after a short number of years that I feel like I must be missing something. How could it be otherwise, given the current system? This is all completely setting aside the developmental limitations of small children.
Let’s take two kids, Jamie and Alex. Pretend that there are no developmental limitations on children’s brains and that they can be taught to read equally well at ages 3, 4, 5, and 6.
Alex starts preschool at age 3 and they can read at a 1st grade level by the time they enter Kindergarten.
Jamie does not do any preschool and cannot read at all when they enter Kindergarten.
By the end of Kindergarten, Alex can read slightly better than 1st grade level, but not a lot better, since the curriculum hasn’t been challenging. It’s basically been a rehash of what they already can do. Jamie can read at the expected grade level by the end of Kindergarten.
By the end of first grade, no accommodations have been made for the fact that Alex is a slightly advanced reader. Both kids are given essentially the same pool of books to read. Alex has not skipped a grade or put in an some secret fast-track program for kids who went to preschool, because this does not exist. So by the end of first grade, they can read about equally well. Maybe Alex reads slightly better, but since no real pressure is being put on this advantage that would cause it to compound rather than diminish, it naturally diminishes until both students are at the same level.
Acting as though anything else would happen doesn’t make sense to me. It’s not like each year a child spends in school exerts some kind of Education Force on their brain which accrues generalized scholastic ability. Kindergartners are taught kindergarten level math and reading skills; kids entering kindergarten who already possess these skills only benefit until the other kids catch up.
So IMO the problem isn’t the preschool “doesn’t do anything”. The problem is that the system as it stands doesn’t actually utilize the potential advantage of preschool. We are pretty far away from a system that would do so; such a system would need to be one that tailors the specific educational content to the specific child.
My four year old can read pretty well and can write well enough that you can puzzle out what he’s trying to communicate. But there is no expectation that he’s going to skip kindergarten because of this. So in what sense could this ever be a long-term academic advantage?
I sometimes worry that ideas are prematurely rejected because they are not guaranteed to work, rather than because they are guaranteed not to work. In the end it might turn out that zero ideas are actually guaranteed to work and thus we are left with an assortment of not guaranteed to work ideas which are underdeveloped because some possible failure mode was found and thus the idea was abandoned early.