4 ms·

Neither will the vast majority of humans for any reasonable length of time unless you apply significant amount of force and incentives for staying on task. If

by vidarh 3y ago

Neither will the vast majority of humans for any reasonable length of time unless you apply significant amount of force and incentives for staying on task. If you want something that will "mindlessly" execute repetitive steps, anything trained to try to act in any way like how humans do would seem an exceedingly bad choice.

imtringued 3y ago
Yeah sure, but most CS students will have to do a simple paper trace of a TM or design a TM diagram to graduate. It was on the exam and they only gave you 90 minutes, so it is clearly possible to train a human to do it. Regarding the second half. You aren't very creative. Sometimes you have something very basic, like two tables with numbers in the columns and you ask the LLM to select table rows using it's language capabilities and then tell it to give you the sum and it doesn't work. Great. Except in two years someone will figure out how to train a model to do this and nobody will care. The bitter lesson doesn't mean architectural changes are worthless, it just tells you that an approach that lets the machine leverage its computational capabilities will outpace a researcher's effort to patch a fundamentally broken model with a thousand small workarounds.
- vidarh 3y ago
  So incentives, in other words. It is possible to teach a human there are consequences not to, and rewards for doing so, because we are consistently fine tuning and retaining state over time, and we have feedback mechanisms in the form of drives. Nothing I said is to suggest it can't be done with LLMs - I'm on record in HN saying LLMs can be made to work as turing machines just fine, and I stand by that. What you can't expect is to give it a prompt, no ongoing constraints or reinforcement, or other feedback, and stay on task for prolonged periods of time. Humans don't do that with any consistency. We self regulate through a whole array of feedback systems, including internal ongoing thoughts, that by the time of exams have been conditioned over years to get better at sticking to a task and consider exam outcomes to matter. And still a lot of us struggle to keep focus for even that kind of time period. Or engage in activities we know are bad for us. Provide similar mechanisms to regular LLMs and they can stay on task too - we can trivially get LLMs to execute individual steps of turing machines, so getting them to keep doing it is a question of how to regulate their ongoing behaviour, not an inherent limitation.
zer00eyz 3y ago
We got to the moon, we built the atomic bomb on the literal backs of calculators. These were mostly women who sat at desks all day long and did math. https://www.smithsonianmag.com/science-nature/history-human-computers-180972202/ https://www.smithsonianmag.com/science-nature/history-human-... It was a pretty good job, there wasn't a lot of force, and money is a decent incentive to stay on task.
- HPsquared 3y ago
  There would have been a system of checking the results somehow. Perhaps the same could be done with LLMs.
  - zer00eyz 3y ago
    More agents is all you need: https://news.ycombinator.com/item?id=39955725 https://news.ycombinator.com/item?id=39955725 This works because each trip through the LLM with the same question uses a different random seed (temperature). You're averaging randomness through a statistical model. You're going to end up with a repeatable result. You're not going to end up with an accurate result.
    - vidarh 3y ago
      You don't typically get accurate results with most humans either without processes for cross checking/tests.
- cjk2 3y ago
  Just a nit pick, we got to the moon also on the back of slide rules and 3 significant figures of pencil scribblings by thousands of shirted engineers. The computer and calculator side bit is somewhat over-explained because it was sexier. Also, my mother was a calculator once. In the UK. For petrochemical company. It was really boring so she went to smoke weed and fix radios instead as it was the 60s by then and that was cooler.
- roenxi 3y ago
  Sure; but an LLM can do all those things too. We're talking about a very restricted problem domain with clear deterministic inputs and outputs, neural nets outperform humans under those conditions. Any computer-based method can. So if we take as a starting point that LLMs can't simulate Turing machines and that is important, the tasks you are referring to won't involve simulating Turing machines either. I personally suspect LLMs can simulate Turing machines. Turing machines are pretty basic. Particularly after noting that humans use external tools like pen & paper, so it'd be fair to train the LLM to use some external memory-aiding tool. Then it becomes an academically trivial problem for even a basic neural net.
criddell 3y ago
I was thinking the same thing. I get that it’s interesting to study what an LLM in isolation can do. That doesn’t accurately reflect what the commercially available AIs are, does it? Aren’t ChatGPT and Gemini AI tool suites? It kind of annoys me that when you ask Copilot for help with something, it will spit out non-working code. Why isn’t Copilot running and debugging the code before presenting it as an answer?

by vidarh 3y ago

imtringued 3y ago

Yeah sure, but most CS students will have to do a simple paper trace of a TM or design a TM diagram to graduate. It was on the exam and they only gave you 90 minutes, so it is clearly possible to train a human to do it. Regarding the second half. You aren't very creative. Sometimes you have something very basic, like two tables with numbers in the columns and you ask the LLM to select table rows using it's language capabilities and then tell it to give you the sum and it doesn't work. Great. Except in two years someone will figure out how to train a model to do this and nobody will care. The bitter lesson doesn't mean architectural changes are worthless, it just tells you that an approach that lets the machine leverage its computational capabilities will outpace a researcher's effort to patch a fundamentally broken model with a thousand small workarounds.

vidarh 3y ago
So incentives, in other words. It is possible to teach a human there are consequences not to, and rewards for doing so, because we are consistently fine tuning and retaining state over time, and we have feedback mechanisms in the form of drives. Nothing I said is to suggest it can't be done with LLMs - I'm on record in HN saying LLMs can be made to work as turing machines just fine, and I stand by that. What you can't expect is to give it a prompt, no ongoing constraints or reinforcement, or other feedback, and stay on task for prolonged periods of time. Humans don't do that with any consistency. We self regulate through a whole array of feedback systems, including internal ongoing thoughts, that by the time of exams have been conditioned over years to get better at sticking to a task and consider exam outcomes to matter. And still a lot of us struggle to keep focus for even that kind of time period. Or engage in activities we know are bad for us. Provide similar mechanisms to regular LLMs and they can stay on task too - we can trivially get LLMs to execute individual steps of turing machines, so getting them to keep doing it is a question of how to regulate their ongoing behaviour, not an inherent limitation.

zer00eyz 3y ago

We got to the moon, we built the atomic bomb on the literal backs of calculators. These were mostly women who sat at desks all day long and did math. https://www.smithsonianmag.com/science-nature/history-human-computers-180972202/ https://www.smithsonianmag.com/science-nature/history-human-... It was a pretty good job, there wasn't a lot of force, and money is a decent incentive to stay on task.

HPsquared 3y ago
There would have been a system of checking the results somehow. Perhaps the same could be done with LLMs.
- zer00eyz 3y ago
  More agents is all you need: https://news.ycombinator.com/item?id=39955725 https://news.ycombinator.com/item?id=39955725 This works because each trip through the LLM with the same question uses a different random seed (temperature). You're averaging randomness through a statistical model. You're going to end up with a repeatable result. You're not going to end up with an accurate result.
  - vidarh 3y ago
    You don't typically get accurate results with most humans either without processes for cross checking/tests.
cjk2 3y ago
Just a nit pick, we got to the moon also on the back of slide rules and 3 significant figures of pencil scribblings by thousands of shirted engineers. The computer and calculator side bit is somewhat over-explained because it was sexier. Also, my mother was a calculator once. In the UK. For petrochemical company. It was really boring so she went to smoke weed and fix radios instead as it was the 60s by then and that was cooler.
roenxi 3y ago
Sure; but an LLM can do all those things too. We're talking about a very restricted problem domain with clear deterministic inputs and outputs, neural nets outperform humans under those conditions. Any computer-based method can. So if we take as a starting point that LLMs can't simulate Turing machines and that is important, the tasks you are referring to won't involve simulating Turing machines either. I personally suspect LLMs can simulate Turing machines. Turing machines are pretty basic. Particularly after noting that humans use external tools like pen & paper, so it'd be fair to train the LLM to use some external memory-aiding tool. Then it becomes an academically trivial problem for even a basic neural net.

criddell 3y ago

I was thinking the same thing. I get that it’s interesting to study what an LLM in isolation can do. That doesn’t accurately reflect what the commercially available AIs are, does it? Aren’t ChatGPT and Gemini AI tool suites? It kind of annoys me that when you ask Copilot for help with something, it will spit out non-working code. Why isn’t Copilot running and debugging the code before presenting it as an answer?