651–660 of 1001 posts
This reminds me that, with our species, the real power always lies with those that tell the better story. Skynet and HAL 9000 had a more profound impact on those researchers than years of practice and study in the field. No surprise therefore that the storytelling of someone like Trump is practically indelibly imprinted into the mind of his supporters.
I was talking to my (12 year old) son about parts of math he finds boring. He said that he thinks absolute value is absurdly easy and extremely boring. I asked him if there was anything that might make it more interesting, he said "maybe complex numbers". So I asked him "what would the absolute value of i+1 be?" he thinks for a little bit and says "square root of 2" and I ask him "what about the absolute value of 2i…
Put this on my tombstone after the robots kill me or whatever, but I think all “AI safety” concerns are a wild overreaction totally out of proportion to the actual capabilities of these models. I just haven’t seen anything in the past year which makes me remotely fearful about the future of humanity, including both our continued existence and our continued employment.
Apparently what made this person fearful was grade school math. "Though only performing maths on the level of grade-school students, acing such tests made researchers very optimistic about Q*’s future success, the source said."
I feel very comfortable saying, as a mathematician, that the ability to solve grade school maths problems would not be at all a predictor of ability to solve real mathematical problems at a research level. The reason LLMs fail at solving mathematical problems is because: 1) they are terrible at arithmetic, 2) they are terrible at algebra, but most importantly, 3) they are terrible at complex reasoning (more specifica…
1. "I don't think this is real news / important because solving grade school math is not a predictor of ability to do complex reasoning."
2. "LLMs can't solve grade school math because they're bad at arithmetic, algebra and most importantly reasoning."
So... from 2 automatically follows that LLMs with sufficiently better math may be sufficiently better at reasoning as you said "most importantly" reasoning is relevant for their ability to do math. Saying "most importantly reasoning" and then saying that reasoning is irrelevant if they can do math, is odd.
Remember, about a month ago Sam posted a comment along the lines of "AI will be capable of superhuman persuasion well before it is superhuman at general intelligence, which may lead to very strange outcomes". The board was likely spooked by the recent breakthroughs (which were most likely achieved by combining transformers with another approach), and hit the panic button. Anything capable of "superhuman persuasion",…
Earlier quoted context omitted.
There are thousands of structures and substances in a human head besides neurons, at all sorts of commingling and overlapping scales, and the neurons in those heads behave much differently and with tremendously more complexity than the metaphorical ones in a neural network. And in a human, all those structures and substances, along with the tens of thousands more throughout the rest of the body, are collectively read…
True. But a human neuron is more complex than an AI neuron by a constant factor. And we can improve constants. Also you say years like it's a lot of data--but they can run RL on chatgpt outputs if they want, isn't it comparable? But anyway i share your admiration for the biological thinking machines ;)
constant still can be not reachable yet: like 100T neurons in brain vs 100B in chatgpt, and also brain can involve some quantum mechanics for example, which will make complexity diff not constant, but say exponential.
Earlier quoted context omitted.
There are thousands of structures and substances in a human head besides neurons, at all sorts of commingling and overlapping scales, and the neurons in those heads behave much differently and with tremendously more complexity than the metaphorical ones in a neural network. And in a human, all those structures and substances, along with the tens of thousands more throughout the rest of the body, are collectively read…
True. But a human neuron is more complex than an AI neuron by a constant factor. And we can improve constants. Also you say years like it's a lot of data--but they can run RL on chatgpt outputs if they want, isn't it comparable? But anyway i share your admiration for the biological thinking machines ;)
I was talking to my (12 year old) son about parts of math he finds boring. He said that he thinks absolute value is absurdly easy and extremely boring. I asked him if there was anything that might make it more interesting, he said "maybe complex numbers". So I asked him "what would the absolute value of i+1 be?" he thinks for a little bit and says "square root of 2" and I ask him "what about the absolute value of 2i…
Earlier quoted context omitted.
Sure, but people have already applied deep learning techniques to theorem proving. There are some impressive results (which the press doesn't seem at all interested in because it doesn't have ChatGPT in the title). It's really harder than one might imagine to develop a system which is good at higher order logic, premise selection, backtracking, algebraic manipulation, arithmetic, conjecturing, pattern recognition, vi…
You seem knowledgeable. Can you share a couple of interesting papers for theorem proving that came out in the last year? I read a few of them as they came out, and it seemed neural nets can advance the field by mixing "soft" language with "hard" symbolic systems.
The most interesting papers to me personally are the following three:
* Making higher order superposition work. https://doi.org/10.1007/978-3-030-79876-5_24
* MizAR 60 for Mizar 50. https://doi.org/10.48550/arXiv.2303.06686
* Magnus Hammer, a Transformer Based Approach to Premise Selection. https://doi.org/10.48550/arXiv.2303.04488
Your mileage may vary.