johnlawrenceaspden

Karma: 1,666

johnlawrenceaspden 16 Sep 2023 10:56 UTC
−7 points
in reply to: [DEACTIVATED] Duncan Sabien’s comment on: Lies, Damn Lies, and Fabricated Options
Haven’t you just “fabricated an option” where it’s possible to talk about politics on less wrong without it turning into a mind-killed clusterfuck? I mean yes, it would be lovely.....

johnlawrenceaspden 28 Jul 2023 22:10 UTC
8 points
−1
on: Semaglutide and Muscle
I wouldn’t touch this stuff with someone else’s bargepole. It looks like it takes the willpower out of starvation, and as the saying goes, you can starve yourself thin, but you can’t starve yourself healthy.
I could be convinced, by many years of safety data and a well understood causal mechanism for both obesity and the action of these drugs, that that’s wrong and that they really are a panacea. But I am certainly not currently convinced!
The question that needs answering about obesity is ‘why on earth are people with enormous excess fat reserves feeling hungry?‘. It’s like having a car with the boot full of petrol in jerry cans but the ‘fuel low’ light is blinking.

johnlawrenceaspden 13 Jul 2023 20:41 UTC
2 points
0
in reply to: Aiyen’s comment on: When do “brains beat brawn” in Chess? An experiment
depends on facts about physics and psychology
It does, and a superintelligence will understand those facts better than we do.

johnlawrenceaspden 4 Jul 2023 12:02 UTC
2 points
0
on: Inference Speed is Not Unbounded
My basic argument is that the there are probably mathematical limits on how fast it is possible to learn.
Doubtless there are! And limits to how much it is possible to learn from given data.
But I think they’re surprisingly high, compared to how fast humans and other animals can do it.
There are theoretical limits to how fast you can multiply numbers, given a certain amount of processor power, but that doesn’t mean that I’d back the entirety of human civilization to beat a ZX81 in a multiplication contest.
What you need to explain is why learning algorithms are a ‘different sort of thing’ to multiplication algorithms.
Maybe our brains are specialized to learning the sorts of things that came in handy when we were animals.
But I’d be a bit surprised if they were specialized to abstract reasoning or making scientific inferences.

johnlawrenceaspden 4 Jul 2023 11:40 UTC
3 points
1
on: Inference Speed is Not Unbounded
All of RL’s successes, even the huge ones like AlphaGo (which beat the world champion at Go) or its successors, were not easy to train. For one thing, the process was very unstable and very sensitive to slight mistakes. The networks had to be designed with inductive biases specifically tuned to each problem.
And the end result was that there was no generalization. Every problem required you to rethink your approach from scratch. And an AI that mastered one task wouldn’t necessarily learn another one any faster.
I had the distinct impression that AlphaZero (the version of AlphaGo where they removed all the tweaks) could be left alone for an afternoon with the rules of almost any game in the same class as go, chess, shogi, checkers, noughts-and-crosses, connect four, othello etc, and teach itself up to superhuman performance.
In the case of chess, that involved rediscovering something like 400 years of human chess theorizing, to become the strongest player in history including better than all previous hand-constructed chess programs.
In the case of go, I am told that it not only rediscovered a whole 2000 year history of go theory, but added previously undiscovered strategies. “Like getting a textbook from the future”, is a quote I have heard.
That strikes me as neither slow nor ungeneral.
And there was enough information in the AlphaZero paper that it was replicated and improved on by the LeelaChessZero open-source project, so I don’t think there can have been that many special tweaks needed?

johnlawrenceaspden 4 Jul 2023 11:00 UTC
3 points
0
on: Could a superintelligence deduce general relativity from a falling apple? An investigation
This is great. Strong upvote!
Are you claiming that a physically plausible superintelligence couldn’t infer the physical laws from a video, or that AIXI couldn’t?
Those seem to be different claims and I wonder which of the two you’re aiming at?

johnlawrenceaspden 3 Jul 2023 22:34 UTC
1 point
0
in reply to: Herb Ingram’s comment on: When do “brains beat brawn” in Chess? An experiment
For example, you might be much smarter than me and a meteorologist, but you’d find it hard to predict the weather in a year’s time better than me if it’s a single-shot-contest.
Sure, but I’d presumably be quite a lot better at predicting the weather in two days time.

johnlawrenceaspden 30 Jun 2023 13:56 UTC
3 points
0
on: When do “brains beat brawn” in Chess? An experiment
I think this is a great article, and the thesis is true.
The question is, how much intelligence is worth how much material?
Humans are so very slow and stupid compared to what is possible, and the world so complex and capable of surprising behaviour, that my intuition is that even a very modest intelligence advantage would be enough to win from almost any starting position.
You can bet your arse that any AI worthy of the name will act nice until it’s already in a winning position.
I would.

johnlawrenceaspden 30 Jun 2023 13:44 UTC
16 points
1
in reply to: Lucius Bushnaq’s comment on: When do “brains beat brawn” in Chess? An experiment
If there’s some intelligence threshold past which minds pretty much always draw against each other in chess even if there is a giant intelligence gap between them, I wouldn’t be that surprised.
Just reinforcing this point. Chess is probably a draw for the same reason Noughts-and-crosses is.
Grandmaster chess is pretty drawish. Computer chess is very drawish. Some people think that computer chess players are already near the standard where they could draw against God.
Noughts-and-crosses is a very simple game and can be formally solved by hand. Chess is only a bit less simple, even though it’s probably beyond actual formal solution.
The general Game of Life is so very far beyond human capability that even a small intelligence advantage is probably decisive.

johnlawrenceaspden 30 Jun 2023 13:24 UTC
5 points
3
in reply to: green_leaf’s comment on: When do “brains beat brawn” in Chess? An experiment
The “purpose” of most martial arts is to defeat other martial artists of roughly the same skill level, within the rules of the given martial art.
Optimizing for that is not the same as optimizing for general fighting. If you spent your time on the latter, you’d be less good at the former.
“Beginner’s luck” is a thing in almost all games. It’s usually what happens when someone tries a strategy so weird that the better player doesn’t immediately understand what’s going on.
The other day a low-rated chess player did something so weird in his opening that I didn’t see the threat, and he managed to take one of my rooks.
That particular trap won’t work on me again, and might not have worked the first time if I’d been playing someone I was more wary of.
I did eventually manage to recover and win, but it was very close, very fun, and I shook his hand wholeheartedly afterwards.
Every other game we’ve played I’ve just crushed him without effort.
About a year ago I lost in five moves to someone who tried the “Patzer Attack”. Which wouldn’t work on most beginners. The first time I’d ever seen it. It worked once. It will never work on me again.

johnlawrenceaspden 30 Jun 2023 13:08 UTC
16 points
0
in reply to: gwern’s comment on: When do “brains beat brawn” in Chess? An experiment
For a clear example of this, in endgames where I have a winning position but have little to no idea how to win, Stockfish’s king will often head for the hills, in order to delay the coming mate as long as theoretically possible.
Making my win very easy because the computer’s king isn’t around to help out in defence.
This is not a theoretical difficulty! It makes it very difficult to practise endgames against the computer.

johnlawrenceaspden 13 Jun 2023 15:27 UTC
14 points
3
in reply to: paulfchristiano’s comment on: Cosmopolitan values don’t come free
Paul, this is very thought provoking, and has caused me to update a little. But:
I loathe factory-farming, and I would spend a large fraction of my own resources to end it, if I could.
I believe that makes me unusually kind by human standards, and by your definition.
I like chickens, and I wish them well.
And yet I would not bat an eyelid at the thought of a future with no chickens in it.
I would not think that a perfect world could be improved by adding chickens.
And I would not trade a single happy human soul for an infinity of happy chickens.
I think that your single known example is not as benevolent as you think.

johnlawrenceaspden 2 Jun 2023 14:17 UTC
2 points
0
in reply to: the gears to ascension’s comment on: AI-kills-everyone scenarios require robotic infrastructure, but not necessarily nanotech
zero-days are a thing, and hell, it’s even possible that there are computers connected to the internet somewhere that don’t get their patches in a timely manner.

johnlawrenceaspden 2 Jun 2023 14:11 UTC
3 points
−2
in reply to: angmoh’s comment on: Experimental Fat Loss
I’m a complete innocent in all this. I’ve never needed to lose weight before, hence appealing for help here. And I don’t know anything about Vilhjálmur Stefánsson or ketogenic diets in general.
I do know that sloth and gluttony aren’t the explanation, because I have been a slothful glutton for most of my life and I never gained much weight, nor lost it in the long periods when I was a sporty glutton. That’s gone wrong recently, hence my search for reasons and techniques.
Wikipedia seems to imply that Vilhjálmur Stefánsson was interested in eskimo-style all-meat diets.
exfatloss seems to be deliberately holding the amount of protein low, and that does seem to be a load-bearing part of his approach. Also the anti-polyunsaturated fats bit, which I find intriguing because it’s such a good theory, and yet it makes predictions which don’t seem to be true.
Ultimately the new GLP-1 agonist weightloss drugs.....
I’m irritatingly fat, not dying of morbid obesity. I wouldn’t touch such things with someone else’s bargepole, absent twenty years of widespread use and researchers motivated to find the unintended consequences.
Chemical patches as a remedy for chemical poisoning is ok, if it’s the best we can do, but unless the problem is some permanent environmental contaminant, I’m sure we can do better than that!

Experimental Fat Loss

johnlawrenceaspden1 Jun 2023 20:26 UTC

23 points

4 comments1 min readLW link

johnlawrenceaspden 23 May 2023 14:16 UTC
2 points
0
on: I don’t want to talk about AI
That seems perfectly reasonable! Enjoy the sunshine while you still can.

johnlawrenceaspden 15 Apr 2023 19:28 UTC
14 points
20
on: [linkpost] Elon Musk plans AI start-up to rival OpenAI
Wonderful news! Good old Elon! I was worried that there might be some last fragment of hope lying around uncrushed.

johnlawrenceaspden 3 Apr 2023 22:38 UTC
2 points
1
in reply to: avturchin’s comment on: AI-kills-everyone scenarios require robotic infrastructure, but not necessarily nanotech
All the computers on my boat are solar-powered. It’ll be fine....

johnlawrenceaspden 3 Apr 2023 17:05 UTC
0 points
−1
on: AI-kills-everyone scenarios require robotic infrastructure, but not necessarily nanotech
You’re an AI. You’ve got an internet connection. You can make copies of yourself. You’re capable of reading instructions for how to take over insecure computers.
Within about a minute you’re on almost every computer in the world.
Fight that.

johnlawrenceaspden 30 Mar 2023 21:31 UTC
3 points
0
in reply to: avturchin’s comment on: Pausing AI Developments Isn’t Enough. We Need to Shut it All Down by Eliezer Yudkowsky
<nogenies>
Yeah, wouldn’t it be great if there was some way to not have a nuclear war or build AI or have everyone die of bird flu?
</nogenies>
I think I know how this game goes.

johnlawrenceaspden

Ex­per­i­men­tal Fat Loss

Experimental Fat Loss