Posts
Comments
I like corner brackets more!
so to quantify sentence S, i prefer ur suggestion "I think it'll rain tomorrow". the percentage is supposed to modify "I think" anyway, so it makes more sense to make them adjacent. it's just more work bc it's novel syntax, but that's temporary.
The principles you propose make a lot of sense! Dropping "I think" or "My best guess" is then for the best.
Also, the underset/underbraces stuff is promising but too much to spend weirdness points on.
See also drossbucket 2018.
Better at getting specific information out of a present piece of information (e.g. becoming skilled at skimming), better at putting tasks aside when they need some time to be processed in the background.
I agree that if at all there is a counternorm to that, and also with the observation that such comments are often (sadly) ignored.
Yeah, there's all these damn confounders :-|
I looks like movies have been getting longer:
Although there seems to have been a slight dip/stagnation in the 2000s and 2010s, and the text concludes with
In conclusion, our intuition was wrong. There is no trend in the movies runtime. The differences are too small to be noticed. We can say that for the last 60 years movies on average have the same length. No matter what criteria we take into account, the result is the same.
I can deliver, it just takes a while.
Sorry @RomanHauksson! Reversed the downvote. I didn't realize this was a joke 🤦
Mark 2023 looks at this, I've summarized the relevant stuff at the end of this section, and the time spent per task has indeed been declining. I don't know about research looking at task classification, but that would be interesting to do.
My current take is that this provides medium evidence—but I think this could also be evidence of higher ability at priorization.
I found this comment both unhelpful and anti-scholarship, and have therefore strong - downvoted.
Maybe to clarify, by question decomposition I mean techniques such as saying " will happen if and only if and and ... all happen, so we estimate and and &c, and then multiply them together to estimate …", which is how it is done in the sources I linked.
Do you by chance have links about how this is done in sports betting? I'd be interested in that.
I don't remember anything specific from reading their stuff, but that would of course be useful. Sadly, I haven't been able to find any more recent investigations into decomposition, e.g. Connected Papers for MacGregor 1999 gives nothing worthwhile after 2006 on a first skim, but I'll perhaps look more at it.
Thanks for the clarification!
One issue I see with this plan is that it seems to rely on some mathematics that appear to me to not be fully worked out, e.g. infrabayesianism and «boundaries» (for which I haven't been able to find a full mathematical description), and it looks unclear to me whether they will actually be finished in time, and if they are, whether they lead to algorithms that are efficient enough to be scaled to such an ambitious project.
I disagree with this: LLMs seem capable to be ~equally good at arguing for false and true positions, if you ask them (as evidenced by the many incorrect proofs produced by Galactica) when asked for it.
I like this!
I would reduce the point score of "A simple math proof proves X" from 20 down to 16. As far as I know, there is no literature on how faulty mathematical proofs are, but from personal history and anecdotes from others (I have many times found "bugs" in my proofs (5 points), and someone I know working in formal verification would often update me on the errors they found in published proofs (8 points)). I'd give a higher weight (18 points) to formally verified proofs. Not deducting more proofs because of the curious observation that even if a proof is faulty, the result it shows is usually true (just as neural networks want to learn, mathematical proofs want to be correct).
Additionally, proofs purportedly about real-world phenomena are often importance-hacked and one needs to read the exact result pretty carefully, which is often not done.
- Randomized Controlled Trial (RCT) results: 25 points
- Meta-analysis of multiple RCTs: 23 points
I find it amusing that GPT-4 considers meta-analyses to be worsening the results they attempt to pool together.
I have.
(What is your criterion for "physically fit"?)
See also Leech 2020.
Thanks for the heads up! I'll correct it.
Agreed! I'd have much more to add, but at ~7k words I decided to publish.
Or, in other words,
Corollary: It'd probably be much better if people used proof assistants like Lean or Coq to teach mathematics. There is some overhead, sure, but they have much better feedback loops than normal mathematics.
Oh nice, another post I don't need to write anymore :-D
Some disjointed thoughts on this I had:
Feedback loops can be characterized along at least three axes:
- Speed: How quickly you get feedback from actions you take. Archery has a very fast feedback loop: You shoot an arrow and one or two seconds later you see what the outcome is.
- Noise: How noisy the feedback is. High-frequency trading has fast feedback loops, but they have a lot of noise, and finding the signal is the difficult part.
- Richness: How much information you're getting. Dating is one example: Online dating has extremely poor feedback loops: only a couple of bits (did the other person respond, what did they respond) per interaction, while talking & flirting with people in person has extremely rich feedback (the entire visual+acustic+sensory field (plus perhaps smell? Don't know much about human pheromones))—probably kilobytes per minimal motor-action, and megabytes per second.
Fast & low-noise & rich feedback loops are the best, and improving the feedback loop in any of those dimensions is super valuable.
As an example, forecasting has meh feedback loops: they can be very slow (days at least, but more likely months or years (!)), the feedback is kind of poor (only a few bits per forecast), but at least there's not that much noise (you forecast what the question says, but maybe this is why forecasters really don't like questions resolving on technicalities—the closest thing to noise).
But one can improve the richness of the forecasting feedback loop by writing out ones reasoning, so one can update on the entire chain of thought once the resolution comes. Similarly, programming has much better feedback loops than mathematics, which is why I'd recommend that someone learn programming before math (in general learn things with fast & rich feedback loops earlier and slow & poor ones later).
Also, feedback loops feel to me like they're in the neighbourhood of both flow & addiction? Maybe flow is a feedback loop with a constant or increasing gradient, while addiction is a feedback loop with a decreasing gradient (leading into a local & shallow minimum).
When I started reading the Sequences, I started doing forecasting on Metaculus within 3 months (while still reading them). I think being grounded at that time in actually having to do reasoning with probabilities & receiving feedback in the span of weeks made the experience of reading the Sequences much more lasting to me. I also think that the lack of focus on any rationality verification made it significantly harder to develop an art of rationality. If you have a metric you have something to grind on, even if you abandon it later.
See also Building Blocks of Politics: An Overview of Selectorate Theory (Yoav Ravid, 2021).
I would simply select a Turing machine in which I am the output of the empty program as the basis of the universal distribution.
In case you were wondering: In my personal data I find no relation between recent masturbation and meditation quality, contrary to the claims of many meditative traditions.
Hi, did anything come out of this experiment?
I believe VR/AR are not going to be as big of a deal as smartphones, and not produce >$200 bio. of revenue in 2030.
There are Pareto-improvements: everyone is made better off by their own standards. There are, similarly, Pareto-worsenings: Everyone is made worse off by their own standard, or their welfare is unchanged.
Then there are Kaldor-Hicks improvements, which happen if one e.g. reallocates the available goods so that the better-off could compensate the now worse-off to create a Pareto improvement. This compensation need not occur, it needs to be merely possible.
Now can there be a Kaldor-Hicks-worsening?
The naive version (everyone is worse (or equally well) off, and there is no way of making a single person better off through redistribution) seems too strong, there is probably always a redistribution that gives the available resources to a single agent.
A simple negation of the criteria, then perhaps makes more sense: A change is a Kaldor-Hicks-worsening if and only if everyone is worse (or equally well) off and there is no way of creating a Pareto-improvement through reallocation.
This implies a anti-Kaldor-Hicks-worsening: A change makes everyone worse off, but there is some reallocation that creates a Pareto improvement.
Example: We have a Sadist and a Masochist. The Masochist starts hurting the Sadist, thus creating opportunity cost for them both. Switching the roles creates a Pareto improvement.
That's a good point, I'll keep it in mind.
Ah, I see. In the table is the log score of my pill predictions (-0.6). I made predictions abojt the accuracy of my predictions—confusing perhaps.
Blinding is powerful! Not sure where you get 80% from, do you mean the number of times when I was directionally right in the prediction?
That's a good point. I have, but shelved them for sometime later™. If it were true that withdrawal effects are important, you'd expect the difference in placebo-caffeine scores to drop as the experiment progresses, but I haven't checked yet.
Ah, okay. That makes much more sense.
I am of course heavily biased. I would have counted myself among the 8% of people who were unwilling to change their minds
Excuse me but what the fuck‽ Maybe kudos for being honest about your disposition.
But this is LessWrong! We consider this kind of stance to be deeply flawed and corrosive to figuring out what facts constitute the world. The rest of the post is okay, but I hope that people can see that this is a warning sign.
Does anyone know whether this bounty is still standing?
This podcast does not appear on my feed on Google Podcasts.
The MIRI Research Guide recommends An Introduction to Decision Theory and Game Theory: An Introduction. I have read neither and am simply relaying the recommendation.
Courtesy of the programming language checklist.
So you're proposing a new economic policy. Here's why your policy will not work.
Your policy depends on science/theorizing that:
☐ has been replicated only once
☐ has failed to replicate
☐ for which there exist no replication attempts
☐ was last taken seriously sometime around 1900
☐ requires a dynamic stochastic general equilibrium model with 200 free parameters to perfectly track reality perfectly
Your policy would:
☐ disincentivize good things
☐ incentivize bad things
☐ both
☐ be wildly unpopular, even though you think it's the best thing since sliced bread (it's not)
☐ You seem to think that taking options away from people helps them
Your policy just reinvents
☐ land-value taxes, but worse
☐ universal basic income, but worse
☐ price discrimination, but worse
☐ demand subsidy, but worse
☐ demand subsidy, better, but that's still no excuse
☐ Your policy sneakily redistributes money from poor to rich people
☐ Your policy only works if every country on earth accepts it at the same time
☐ You actually have no idea what failure/success of your policy would look like
You claim it fixes
☐ climate change
☐ godlessness
☐ police violence
☐ wet socks
☐ teenage depression
☐ rising rents
☐ war
☐ falling/rising sperm counts/testosterone levels
You seem to assume that
☐ privatization always works
☐ privatization never works
☐ your country will never become a dictatorship
☐ your country will always stay a dictatorship
the cost of coordination is
☐ negligible
☐ zero
☐ negative
☐ Your policy is a Pareto-worsening
In conclusion,
☐ You have copied and mashed together some good ideas with some mediocre ideas
☐ You have not even tried to understand basic economics/political science/sociology concepts
☐ Living under your policy is an adequate punishment for inventing it
One could introduce 🌵 for such users.
Some people definitely say they believe climate change will kill all humans.
In a current caffeine self-experiment I'm tracking the following variables:
- Meditation performance using Meditavo (unfortunately only exportable using the premium version)
- Mindfulness
- Concentration
- Cognition using Anki flashcards scores, exported from the Collection.anki2 by using sqlite3
- Mood via Moodpatterns
- Happiness, Contentment and Stress
- I use the additional interested/disinterested spectrum for tracking horniness
- Creativity and productivity for the day via a simple self-written script that activates via scron
More here. Maybe you could try to self-blind yourself, e.g. have half of your injections be saline and the other half testosterone? (I don't know much about how testosterone is consumed, maybe the benefits accrue slowly enough that randomization doesn't make sense).
Happy to respond to more questions as well, but I haven't finished my experiment yet so I can't provide any code for analysis.
Historically, LessWrong was seeded by the writings of Eliezer Yudkowsky, an artificial intelligence researcher.
He usually descibes himself as a decision theorist if asked for a description of his job.
Man I think I am providing value to the world by posting and commenting here. If it cost money I would simply stop posting here, and not post anywhere else.
The value flows in both directions. I'm fine not getting paid but paying is sending a signal of "what you do here isn't appreciated".
(Maybe I'd feel different if the money was reimbursed to particularly good posters? But then Goodharts law)
I encourage you to fix the mistake. (I can't guarantee that the fix will be incorporated, but for something this important it's worth a try).
I notice I am confused. How do you violate an axiom (completeness) without behaving in a way that violates completeness? I don't think you need an internal representation.
Elaborating more, I am not sure how you even display a behavior that violates completeness. If you're given a choice between only universe-histories and , and your preferences are imcomplete over them, what do you do? As soon as you reliably act to choose one over the other, for any such pair, you have algorithmically-revealed complete preferences.
If you don't reliably choose one over the other, what do you do then?
- Choose randomly? But then I'd guess you are again Dutch-bookable. And according to which distribution?
- Your choice is undefined? That seems both kinda bad and also Dutch-bookable to me tbh. Alwo don't see the difference between this and random choice (shodt of going up in flames, which would constigute a third, hitherto unassumed option).
- Go away/refuse the trade &c? But this is denying the premise! You only have universe-histories and tp choose between! I think what happens with humans is that they are often incomplete over very low-ranking worlds and are instead searching for policies to find high-ranking worlds while not choosing. I think incomplwteness might be fine if there are two options you can guarantee to avoid, but with adversarial dynamics that becomes more and more difficult.
To clarify the comment for @tjaffee, a superintelligence could do the following
- Use invalid but extremely convincing arguments that make the "doomer"[1] change his mind. This appears realistic because sometimes people become convinced of false things through invalid argumentation[2].
- Give a complete plan for an aligned superintelligence, shutting itself down in the process, creating a true and probably convincing argument (this is vanishingly unlikely).
Nice. Would you mind if I took inspiration from your list (crediting you of course).