I am currently a nuclear engineer with a focus in nuclear plant safety and probabilistic risk assessment. I am also an aspiring EA, interested in X-risk mitigation and the intersection of science and policy.
ErickBall(Erick Ball)
Karma: 347
Totally agree about having weights at home. Besides the cost, one upside is there’s no energy barrier to exercising—I can take a 1-minute break from browsing the web or whatever, do a set, and go back to what I was doing without even breaking a sweat. A downside is it’s harder to get in the mindset of doing a full high-intensity workout for 45 minutes; but I think it’s a good tradeoff overall.
In practice the big difference is that KN95 masks generally have ear loops, while N95 masks have straps that go around the back of your head which makes them fit tighter and seal better against your face. Traditional N95 masks (but not the duckbill type discussed here) also have more structure and are less flexible, which might help with fit depending on your face shape.
I like this as a way to clarify my intuition. But I think (as some other commenters here and on the EA forum pointed out) it would help to extend it to a more realistic example.
So let’s say instead of hearing a commotion by the river as I start my walk, I’m driving somewhere, and I come across a random stranger who was walking next to the road, and a car swerves over into the shoulder and is about to hit him. There’s a fence so the pedestrian has no way to dodge. The only thing I can do is swerve my car into the other car to make it hit the fence and stop; this won’t be dangerous to me or the other driver, but it will wreck both cars. And I’m pretty sure it will work. But my car will cost $5,000 to replace (let’s leave insurance out of this hypothetical). Of course, I’ll do it—the poor guy’s life is at stake.
Then the next time I’m driving somewhere, I see this happen again. Would I do it the second time? I mean, yeah, probably. I technically can replace the car again, though it’ll strain things a bit. But I’m definitely going to start thinking about the other factors involved. Why do so many pedestrians feel like they have to walk on this shoulder? Could nobody build a goddamn sidewalk? And why do the other drivers have such poor steering? They’re obviously not trying to kill pedestrians but something has gone very, very wrong. The third time it happens, I think I’ll keep driving, and start looking for more systematic solutions. Throwing money at the problem is clearly better than nothing, up to a point, but it doesn’t seem like the best possible move.
Some people think that multivitamins are actually harmful (or at least cause harms that partially cancel out the benefits) because they contain large amounts of certain things like manganese that we may already get too much of from food.
Parrot-phrasing comes across as kind of manipulative in this description:
saves you the trouble of thinking of suitable paraphrases.
prevents the distracting and time-consuming disagreements (“That’s not quite what I meant”) which often arise over slight differences in wording.
conceals your lack of knowledge or understanding about a subject. It’s quite hard to make a fool of yourself it you only use the other person’s words!
This is exactly the opposite of curiosity, it’s an attempt to gloss over your ignorance, which seems both lazy and mean to the person you’re talking to.
Hey, do you know if there are any results on the human or animal trials yet? I haven’t been able to find anything, even though it’s been a few months and it seems like initial data ought to be coming in.
#24 made me laugh
I love #24 and #41
Okay well it took me more than an hour to get to 50, but still a great exercise!
1. Chemical rocket
2. Launch off of space elevator beyond geosync
3. Giant balloon (aim carefully you can’t steer)
4. Coilgun
5. Launch loop (seriously how has nobody built this yet)
6. Railgun
7. Nuclear thermal rocket
8. Electric (ion) rocket powered by capacitors or batteries (ok might be a little heavy)
9. Electric (ion) rocket powered by lasers from the ground
10. Ablation rocket powered by lasers from the ground
11. Spaceplane combined with any of the rocket types, especially ablation rocket
12. Giant crossbow
13. One of those extending boxing glove toys with all the struts forming parallelograms (apparently this mechanism is called a “pantograph”)
14. Series of nested giant crossbows: each one shoots a smaller crossbow
15. Hitch a ride on a passing asteroid using a tether
16. Use superconductors to levitate a flux-pinned magnet to the moon
17. Fusion rocket
18. Project Orion-style nuclear bomb rocket
19. Run a tether from the moon to the earth, and just let the tip drag along the ground and you can attach climbers to it as it goes by every day
20. Antimatter rocket
21. Ramjet
22. Bussard ramjet
23. Trebuchet
24. Bubble cavitation in a vacuum (could launch a tiny particle at high velocity from LEO)
25. Light gas gun
26. Plasma gun? Is that a thing?
27. Combination balloon and solar sail. The balloon lifts the solar sail until the air is too thin to keep rising, and then the sail takes off.
28. Build a giant radiotelescope to contact space aliens and ask them to carry something to the moon for you.
29. Ablation rocket powered by the sun (using dry ice maybe)
30. Hack the simulation and add code to teleport stuff on your command
31. Build a space elevator on the moon and use it to launch chunks of rock constantly towards earth, which you catch with a see-saw contraption to launch something smaller back the other way
32. Nuclear explosion underground that launches a big chunk of steel through a borehole, like in Operation Plumbob (have to use something longer and tungsten-coated so it doesn’t vaporize)
33. Light sail powered by lasers from the ground
34. Use lightning to superheat pressurized gas in a massive gun chamber
35. Kite that turns into solar sail once it clears most of the atmosphere
36. Build a giant tower from Earth that reaches almost to the moon, with a vacuum chamber inside so it can be supported by an electron beam, and then just toss things from the top
37-39. Practice, practice, practice
40. Get that arm surgery that lets pitchers throw faster than they did before they were injured, but like 50 times
41. Stow away on the next moon launch
42. Bring the moon to you: rob the Earth of its rotational energy by over-using your space elevator, and the moon will slowly drift closer due to tidal locking (I think)
43. Fake a Dr. Evil-style terrorist threat to force the government to send people to the moon to stop you.
44. Rename your home to “the moon”
45. Something with carbon nanotubes (grow them to the moon I guess?)
46. Great great great great great pyramid
47. Drink way too much coffee in order to come up with more ideas (recursive self-improvement technique?)
48. Nuclear-powered jet engine that builds up momentum by circling the Earth a bunch of times like superman, at slightly higher altitude each time
49. Some kind of reactionless drive based on Hawking radiation
50. Just send neutrinos to the moon, nothing will stop them. (You are already doing this.)
I eventually sent them this article, not ideal but good enough:
I would love to have a link to send my parents to convince them to take Vitamin D as a prophylactic. The one RCT, as noted above, has various issues that make it not ideal for that purpose. Does anyone know of an article (by some sort of expert) that makes a good case for supplementation?
Since the same transformer architecture works on images with basically no modification, I suspect it would do well on audio prediction too. Finding a really broad representative dataset for speech might be difficult, but I guess audiobooks are a good start. The context window might cause problems, because 2000 byte pairs of text takes up a lot more than 4000 bytes in audio form. But I bet it would be able to mimic voices pretty well even with a small window. (edit: Actually probably not, see Gwern’s answer.)
If your question is whether the trained GPT-3 model could be modified to work with audio, I suspect not. In principle there are layers of abstraction that a transformer should be able to take advantage of, so that word prediction is mostly uncoupled from audio processing, but there’s not a perfect separation, and we wouldn’t know how to interface them. Maybe you could train a separate transformer model that just transcribes audio into text, and stitch them together that way, but there’s not much reason to think it would be a big improvement over existing speech recognition systems.
Weapons grade is kind of a nebulous term. In the broadest sense it means anything isotopically pure enough to make a working bomb, and in that sense Little Boy obviously qualifies. However, standard enrichment for later uranium bombs is typically around 90%, and according to Wikipedia, Little Boy was around 80% average enrichment.
It is well known that once you have weapons-grade fissile material, building a crude bomb requires little more than a machine shop. Isotopic enrichment is historically slow and expensive (and hard to hide), but there could certainly be tricks not yet widely known…
I’m a big fan of crowd noises for improving concentration when you need to drown out other voices, especially a TV. Much more effective than other forms of white noise.
I think your formatting with the semicolons and the equals sign has confused the transformer. All the strange words, plurals, and weird possessives may also be confusing. On TTT, if I use common words and switch to colon and linebreak as the separators, it at least picks up that the pattern is gibberish: words.
For example:
kobo: book ntthrgeS: Strength rachi: chair sviion: vision drao: road ntiket: kitten dewdngi: wedding lsptaah: asphalt veon: oven htoetsasu: southeast rdeecno: encoder lsbaap1: phonetics nekmet: chic-lookin’ zhafut: crinkly lvtea: question mark cnhaetn: decorated gelsek: ribbon odrcaa: ribbon nepci: ball plel: half cls: winged redoz: brightness star: town moriub:
It seems like the situation with bridges is roughly analagous to neural networks: the cost has nothing to do with how much you change the design (distance) but instead is proportional to how many times you change the design. Evaluating any change, big or small, requires building a bridge (or more likely, simulating a bridge). So you can’t just take a tiny step in each of n directions, because it would still have n times the cost of taking a step in one direction. E. Coli is actually pretty unusual in that the evaluation is nearly free, but the change in position is expensive.
I like this visualization tool. There are some very interesting things going on here when you look into the details of the network in the second-to-last MNIST figure. One is that it seems to mostly identify each digit by ruling out the others. For instance, the first two dot-product boxes (on the lower left) could be described as “not-a-0” detectors, and will give a positive result if they detect pixels in the center, near the corners, or at the extreme edges. The next two boxes could be loosely called “not-a-9“ detectors (though they also contribute to the 0 and 4 classifications) and the three after that are “not-a-4” detectors. (The use of ReLU makes this functionally different from having a “4” detector with a negative weight.)
Now, take a look at the two boxes that contribute to output 1 (I would call these “not-a-1” detectors). If you select the “7″ input and see how those two boxes respond to it, they both react pretty strongly to the very bottom of the 7 (the fact that it’s near the edge) and that’s what allows the network to distinguish a 7 from a 1. Intuitively, this seems like a shared feature—so why is the network so sure that anything near the bottom cannot be part of a 1?
It looks to me like it’s taking advantage of the way the MNIST images are preprocessed, with each digit’s center-of-mass translated to the center of the image. Because the 7 has more of its mass near the top, its lower extremity can reach farther from the center. The 1, on the other hand, is not top-heavy or bottom-heavy, so it won’t be translated by any significant amount in preprocessing and its extremities can’t get near the edges.
The same thing happens with the “not-a-3” detector box when you select input 2. The “not-a-3″ detector triggers quite strongly because of the tail that stretches out to the right. That area could never be occupied by a 3, because the 3 has most of its pixel mass near its right edge and will be translated left to get its center of mass centered in the image. The “7” detector (an exception to the pattern of ruling digits out) mostly identifies a 7 by the fact that it does not have any pixels near the top of the image (and to a lesser extent, does not have pixels in the lower-right corner).
What does this pattern tell us? First, that a different preprocessing technique (centering a bounding box in the image, for instance, instead of the digit’s center of mass) would require a very different strategy. I don’t know off hand what it would look like—maybe there’s some trick for making the problems equivalent, maybe not. Second, that it can succeed without noticing most of what humans would consider the key features of these digits. For the most part it doesn’t need to know the difference between straight lines and curved lines, or whether the parts connect the way they’re supposed to, or even whether lines are horizontal or vertical. It can use simple cues like how far each digit extends in different directions from its center of mass. Maybe with so few layers (and no convolution) it has to use those simple cues.
As far as interpretability, this seems difficult to generalize to non-visual data, since humans won’t intuitively grasp what’s going on as easily. But it certainly seems worthwhile to explore ideas for how it could work.
I like LearnObit so far. We should talk sometime about possible improvements to the interface. Are you familiar with quantum.country? Similar goal, different methods, possible synergy. They demo what is (in my opinion) a very effective teaching technique, but provide no good method for people to create new material. I think with some minor tweaks LearnObit might be able to fill that gap.
Take a look at the Fasting-Mimicking Diet, which has some decent evidence going for it. It’s a 5-day period of low calorie consumption with restricted carb and protein intake, repeated every few months.
Some people actually think the benefits of caloric restriction (to the extent there are any benefits in humans beyond just avoiding overfat) may result from incidental intermittent fasting. I’m no expert but my fairly vague understanding is that the re-feeding period after a fast promotes some kind of cellular repair process that doesn’t occur if you’re continuously well-fed. I guess people who restrict calories overall would generally get little doses of this every once in a while as their food intake fluctuates by chance.
The Kelly criterion is intended to maximize log wealth. Do you think that’s a good goal to optimize? How would your betting strategy be different if your utility function were closer to linear in wealth (e.g. if you planned to donate most of it above some threshold)?