↘↘↘↘↘↘↙↙↙↙↙↙
Checkout my Biography.
↗↗↗↗↗↗↖↖↖↖↖↖
Johannes C. Mayer
Karma: 1,273
The Insanity Detector and Writing
An extreme test of how much you trust a persons intention is to consider whether you would upload them, if you could. Then they would presumably be the only (speed) superintelligence.
Maybe better name: Let me help debug your math via programming
If you’ve tried this earnestly 3 times, after the 3rd time, I think it’s fine to switch to just trying to solve the level however you want (i.e. moving your character around the screen, experimenting).
After you failed 3 times, wouldn’t it be a better exercise to just play around in the level until you get a new pice of information that you predict will allow you to reformulate better plans, and then step back into planning mode again?
Another one: We manage to solve alignment to a significant extend. The AI who is much smarter than a human thinks that it is aligned, and takes aligned actions. The AI even predicts that it will never become unaligned to humans. However, at some point in the future as the AI naturally unrolles into a reflectively stable equilibrium it becomes unaligned.
Why not AI? Is it that AI alignment is too hard? Or do you think it’s likely one would fall into the “try a bunch of random stuff” paradigm popular in AI, which wouldn’t help much in getting better at solving hard problems?
What do you think about the strategy of instead of learning a textbook e.g. on information theory, or compilers you try to write the textbook and only look at existing material if you are really stuck. That’s my primary learning strategy.
It’s very slow and I probably do it too much, but it allows me to train to solve hard problems that aren’t super hard. If you read all the text books all the practice problems remaining are very hard.
The Legacy of Computer Science
How about we meet, you do research, and I observe, and then try to subtly steer you, ideally such that you learn faster how to do it well. Basically do this, but without it being an interview.
What are some concrete examples of the of research that MIRI insufficiently engaged with? Are there general categories of prior research that you think are most underutilized by alignment researchers?
… and Carol’s thoughts run into a blank wall. In the first few seconds, she sees no toeholds, not even a starting point. And so she reflexively flinches away from that problem, and turns back to some easier problems.
I spend ~10 hours trying to teach people how to think. I sometimes try to intentionally cause this to happen. Usually you can recognize it by them starting to be quiet (I usually give the instruction that they should do all their thinking out loud). And this seems to be when actual cognitive labor is happening, instead of saying things that you already knew. Though usually they by default fail earlier than “realizing the hard parts of ELK”.
Usually I need to tell them that actually they are doing great by thinking about the black wall more, and shouldn’t now switch the topic.
Infact it seem to be a good general idea generation strategy to just write down all the easy ideas first, until you hit this wall, such that you can start to actually think.
Why Physicists are competent
Here is my current model after thinking about this for 30 minutes of why physicists are good at solving hard problems (not ever having studied physics extensively myself).
The job description of a physicist is basically “understand the world”, meaning make models that have predictive power over the real world.
This is very different from math. In some sense a lot harder. In math you know everything. There is no uncertainty. And you have a very good method to verify that you are correct. If you have generated a proof, it’s correct. It’s also different from computer science for similar reasons.
But of cause physicists need to be very skilled at math, because if you are not skilled at math you can’t make good models that have predictive power. Similarly physicists need to be good at computer science, to implement physicsal simulations, which often involve complex algorithms. And to be able to actually implement these algorithms such that they are fast enough, and run at all, they need to also be decent at software engeneering.
Also understanding the scientific method is a lot more important when you are physicist. It’s sort of not required to understand science for doing math and theoretical CS.
Another thing is that physicists need actually do things that work. You can do some random math that’s not useful at all. It seems harder to make a random model of reality that predicts some aspect of reality that you couldn’t predict before, and have you not figure out anything important. As a physicist you are actually measured by how reality is. You can’t go “hmm maybe this just doesn’t work” like in math. Obviously somehow it works because it’s reality, you just haven’t figured out how to properly capture how reality is in your model.
Perhaps this trains physicist to not give up on problems, because the default assumption is that clearly there must be some way to model some part of reality, because reality is in some sense already a model of itself.
I think this is the most important cognitive skill. Not giving up. I think this is much more important than any particular pice of technical knowledge. Having technical knowledge is of cause required, but it seems that if you where to not give up on thinking how to solve a problem (that is hard but important) would make you end up learning whatever is required.
And in some sense it is this simple. When I see people run into a wall, and then have them stare at a wall they often have ideas that I like so much that I feel the need to write them down.
I watched this video, and I semi trust this guy (more than anybody else) about not getting it completely wrong. So you can eat too much soy. But eating a bit is actually healthy, is my current model.
Here is also a calculation I did that it is possible to get all amino acids from soy without eating too much.
Haven’t thought about, nor experimented with that. If you think clams would be ok to eat, you could perform the experiment yourself.
At the 2024 LessWrong Community weekend I met somebody who I have been working with for perhaps 50 hours so far. They are better at certain programming related tasks than me, in a way provided utility. Before meeting them they where not even considering working on AI alignment related things. The conversation wen’t something like this:
Johannes: What are you working on.
Other Person: Web development. What are you working on?
Johannes: I am trying to understand intelligence such that we can build a system that is capable enough to prevent other misaligned AI’s from being build, and that we understand enough such that we can be sure that it wouldn’t kill us. [...] Why are you not working on it? Other Person: (I forgot what he said)
Johannes: Oh then now is the perfect time to start working on it.
Other Person: So what are you actually doing.
Johannes: (Describes some methodologies.)
Other Person: (Questions whether these methodologies are actually good, and thinks about how they could be better.)
[...]Actually this all happened after the event when traveling from the venue to the train station.
It doesn’t happen that often that I get something really good out of a random meeting. Most of them are bad. However, I think the most important thing I do to get something out is to just immediately talk about the things that I am interested in. This efficiently filters out people, either because they are not interested, or because they can’t talk it.
You can overdo this. Starting a conversation with “AI seems very powerful, I think it will likely destroy the world” can make other people feel awkward (I know from experience). However, the above formula of “what do you do” and then “and I do this” get’s to the point very quickly without inducing awkwardness.
Basically you can think of this as making random encounters (like walking back to the train station with randomly sampled people) non-random by always trying to steer any encounter such that it becomes useful.
I probably did it badly. I would eat hole grain bread pretty regularly, but not consistently. I might not eat it for 1 week in a row sometimes. That was before I knew that amino acids are important.
It was ferritin. However the levels where actually barely within acceptable levels. I hypothesise that because I started to eat steamed blood for perhaps 2 weaks prior every day, and that blood contains a lot of heme iron, that I was deficient before.
I think running this experiment is generally worth it. It’s very different to read a study and to run the experiment and see the effect yourself. You may also try to figure out if you are amino acid deficient. See this comment, as well as others in that comment stack.
The reason I mention chicken is that last time I ran this experiment with beef my body started to hurt really bad such that I woke up in the middle of the night. I am pretty sure that the beef was the reason. Maybe something weird was going on in my body at the same time. However, when I tried the same one week later with chicken I didn’t have this issue.
I ate tons of beluga lentils. Sometimes 1kg (cooked) a day. That wasn’t enough. However, now I switched to eating 600g (cooked) soybeans every day, and that was a very significant improvement (like solving the problem to 75% or so). Soy is a complete protein. Soy beans are also very cheap.
Don’t spend all your time compressing knowledge that’s not that useful to begin with, if there are higher value things to be learned.