6 ms·

I know the article focused on the deeper/code-ier aspects of AI. However, I think that it's mundane aspects are going to be huge. I once had a co-researcher th

I know the article focused on the deeper/code-ier aspects of AI. However, I think that it's mundane aspects are going to be huge. I once had a co-researcher that asked for help with his Matlab code. We were in a smell-lab together working on mice. His code was about controlling some scent valves in response to a mouse putting it's nose in a hole and breaking a laser beam. Importantly, this code was running his thesis, essentially, and he had no coding experience prior to this experiment. I said sure, I'd love to help, but you owe me a six-pack. So we sat down one day and started going through his Matlab code. After the 18th nested 'if' statement, I had to up the payment to a case of beer. LLMs would have helped my co-researcher out a lot. He might have actually gotten the code working at least. Most HNers don't appreciate, I think, how difficult coding is most people. AIs are already helping bridge that gap. Another more mundane area is in communication. Research papers are notorious for being obtuse and jargon filled, to the point where many fields are nearly speaking another language entirely. AIs can help with that as well. Not only in the writing side, but also in the reading side. Can can put a paper into them and ask for summaries, you can put multiple papers in and ask for review papers. You must be very careful, of course, but the speed increase is just amazing.

boredtofears 3y ago
How would they know if the code produced by ChatGPT had a side effect that effected their research? Isn't it risky to not fully understand the behavior of your experiment?
- kian 3y ago
  I think in this case that problem even exists in their own code that they don't fully understand.
- _giorgio_ 3y ago
  Breaking down code in cells and asking chatGPT for help is quite easy and mostly foolproof, if you have enough will to learn things.
  - dpflan 3y ago
    Indeed, build up the complexity from smaller, verifiable code components.
- Balgair 3y ago
  I get what you're saying, but the reality is that most researchers aren't any more code literate than the average person. Which is to say, they know just enough to make things worse. Hopefully those errors are removed before publication, but in my experience, they are not. Especially when it comes to coding. Also, a larger point is that this was fundamental research. A very hard thing. You're supposed to be getting it wrong most of the time. Then when you think you've got it right, you try to publish. Then you hope the reviewers actually read anything you wrote and catch all the errors [0]. But a lot of the time, errors sneak through and really only get discovered after a few years/decades. That's fine in the end, as science is a 'strongest-link' system where we really only care about the big and correct discoveries. 'Weak-link' systems are the opposite, they are governed by the small little falsehoods and errors. [0] Very unlikely, reviews almost always only review to see if the paper is of a high enough quality for the journal to include. I want to be clear on this point: Reviewers do not re-do experiments and typically do not go back through your math, let alone the code you used. Now, each field is different and has different standards, and those vary wildly. But generally, reviewers aren't there to make sure the science is right.
  - kiba 3y ago
    It sounds like that researchers should just learn to properly code.
    - sdenton4 3y ago
      If the set theorists knew how to farm, then we wouldn't need the NSF to give them money.
      - kiba 3y ago
        Given that it is relevant to their research, they should consider learning how to code. Otherwise you'll be at the mercy of whatever autogenerated code that you happen to be using, since you literally don't have the skill to check if it's actually working. The other alternative is to work with a programmer.
nitwit005 3y ago
You have to understand the code the AI emits, to be able to check if it's done what you wanted. That's sometimes more challenging than writing it yourself, as it may use features you're not familiar with. Which is not to say it can't be useful, but don't expect a cure all.
- IanCal 3y ago
  What they're describing is dealing with a person who already doesn't understand what their code does.
- mettamage 3y ago
  Reading code is easier than starting from scratch. Not sure how that is for a researcher though, just my SWE perspective
  - sdenton4 3y ago
    They're different kinds of easy... Writing fresh code is easy because you just express your ideas. What's hard is hitting all the contingencies appropriately; it can take years of 'burn in' to get right. Reading code is easy - especially if you don't know the language well - because someone has already assembled all the correct syntax, etc. What's hard is that the code may be convoluted from years of monkey-patches to handle all those inconvenient contingencies, to the point of obscuring the core ideas. eg, making you search through ten nested indirections to figure out how something actually executes... On the writing side, LLM's make it much easier to have the idiomatics of the language easily on hand. (eg, I never ever remember the exact way to use 'super' in python... but ML assistance is happy to fill in the parens as soon as I start them.)
  - bamboozled 3y ago
    Not really, reading existing code is hard because you have to be on guard that your bias isn’t skewing your judgement about what’s going on. What I’m getting from these comments is (as expected), most people are happy to take the easy assumption filled route rather than actually applying rigour.
  - 3y ago
credit_guy 3y ago
> LLMs would have helped my co-researcher out a lot. This describes me quite well. During the last few days I was working on getting the matrix for an integral operator, using Chebyshev polynomials. Numpy has regular and Chebyshev polynomials, but the documentation is a bit of a mess. I asked ChatGPT for help. It didn't work that well. But it got me started. It took a bit, because Numpy has actually 2 APIs for all their polynomials, and the ChatGPT code was mixing them. Once I figured that out, I decided to stick to a single API, and I finished the code snipped in about half an hour. Without ChatGPT it would have been much more difficult. How many people out there are aware of the Chebyshev polynomials in numpy? Certainly nobody in my circle. ChatGPT was aware, but it was also hallucinating quite a bit. Still, it helped me get the job done.

boredtofears 3y ago

How would they know if the code produced by ChatGPT had a side effect that effected their research? Isn't it risky to not fully understand the behavior of your experiment?

kian 3y ago
I think in this case that problem even exists in their own code that they don't fully understand.
_giorgio_ 3y ago
Breaking down code in cells and asking chatGPT for help is quite easy and mostly foolproof, if you have enough will to learn things.
- dpflan 3y ago
  Indeed, build up the complexity from smaller, verifiable code components.
Balgair 3y ago
I get what you're saying, but the reality is that most researchers aren't any more code literate than the average person. Which is to say, they know just enough to make things worse. Hopefully those errors are removed before publication, but in my experience, they are not. Especially when it comes to coding. Also, a larger point is that this was fundamental research. A very hard thing. You're supposed to be getting it wrong most of the time. Then when you think you've got it right, you try to publish. Then you hope the reviewers actually read anything you wrote and catch all the errors [0]. But a lot of the time, errors sneak through and really only get discovered after a few years/decades. That's fine in the end, as science is a 'strongest-link' system where we really only care about the big and correct discoveries. 'Weak-link' systems are the opposite, they are governed by the small little falsehoods and errors. [0] Very unlikely, reviews almost always only review to see if the paper is of a high enough quality for the journal to include. I want to be clear on this point: Reviewers do not re-do experiments and typically do not go back through your math, let alone the code you used. Now, each field is different and has different standards, and those vary wildly. But generally, reviewers aren't there to make sure the science is right.
- kiba 3y ago
  It sounds like that researchers should just learn to properly code.
  - sdenton4 3y ago
    If the set theorists knew how to farm, then we wouldn't need the NSF to give them money.
    - kiba 3y ago
      Given that it is relevant to their research, they should consider learning how to code. Otherwise you'll be at the mercy of whatever autogenerated code that you happen to be using, since you literally don't have the skill to check if it's actually working. The other alternative is to work with a programmer.

nitwit005 3y ago

You have to understand the code the AI emits, to be able to check if it's done what you wanted. That's sometimes more challenging than writing it yourself, as it may use features you're not familiar with. Which is not to say it can't be useful, but don't expect a cure all.

IanCal 3y ago
What they're describing is dealing with a person who already doesn't understand what their code does.
mettamage 3y ago
Reading code is easier than starting from scratch. Not sure how that is for a researcher though, just my SWE perspective
- sdenton4 3y ago
  They're different kinds of easy... Writing fresh code is easy because you just express your ideas. What's hard is hitting all the contingencies appropriately; it can take years of 'burn in' to get right. Reading code is easy - especially if you don't know the language well - because someone has already assembled all the correct syntax, etc. What's hard is that the code may be convoluted from years of monkey-patches to handle all those inconvenient contingencies, to the point of obscuring the core ideas. eg, making you search through ten nested indirections to figure out how something actually executes... On the writing side, LLM's make it much easier to have the idiomatics of the language easily on hand. (eg, I never ever remember the exact way to use 'super' in python... but ML assistance is happy to fill in the parens as soon as I start them.)
- bamboozled 3y ago
  Not really, reading existing code is hard because you have to be on guard that your bias isn’t skewing your judgement about what’s going on. What I’m getting from these comments is (as expected), most people are happy to take the easy assumption filled route rather than actually applying rigour.
- 3y ago

credit_guy 3y ago

> LLMs would have helped my co-researcher out a lot. This describes me quite well. During the last few days I was working on getting the matrix for an integral operator, using Chebyshev polynomials. Numpy has regular and Chebyshev polynomials, but the documentation is a bit of a mess. I asked ChatGPT for help. It didn't work that well. But it got me started. It took a bit, because Numpy has actually 2 APIs for all their polynomials, and the ChatGPT code was mixing them. Once I figured that out, I decided to stick to a single API, and I finished the code snipped in about half an hour. Without ChatGPT it would have been much more difficult. How many people out there are aware of the Chebyshev polynomials in numpy? Certainly nobody in my circle. ChatGPT was aware, but it was also hallucinating quite a bit. Still, it helped me get the job done.