3 ms·
Magic is that which can not be explained. Could you explain GPT-4's results if you saw them a few years ago?
by esafak 4y ago
Magic is that which can not be explained. Could you explain GPT-4's results if you saw them a few years ago?
- Jtsummers 4y agoProbably, I mean I first studied ANNs over two decades ago and had conversations with people about them prior to starting college in the 90s who had developed solutions with them in the 80s (obviously severely computationally constrained in those days compared to today). So still not magic. To be very blunt: If you believe that GPT and the like are magic, then you're not thinking clearly. You're blinded by the results (which are impressive).
- esafak 4y agoANNs in the 90s were nothing like this. They were not even like this a few years ago. And neither were HMMs. There is an emergent human quality to them because they have approached our abilities; the comparison is tenable, whereas before it was not.
- Jtsummers 4y agoScale and topology. That's the difference between ANNs of the 90s and 00s and today. They're still based on the same fundamental principles and doing (modulo scale) the same fundamental things: classification, prediction, generation. It is not magic, it has never been magic.
- esafak 4y agoScale and topology also differentiates all living creatures. In fact, our topologies are more similar than that of various neural networks, due to evolution. "Classification, prediction, generation" encapsulates everything we do too. So I guess we are not magical either.
- Jtsummers 4y ago> So I guess we are not magical either. We aren't, and I haven't said otherwise.
- esafak 4y agoWe have a different understanding of magic. Say if someone pressed a button and a human-like thing emerged out of a machine, I would call that pretty magical. Even if it was DNA-based, which we "understand", or ran an ML model, which we "understand". This is something that never come close to being done. Yet I think you would not find it magical. Einstein found wonder in the simplicity of a circle. What do you find magical?
- Jtsummers 4y agoYou initially wrote: > Magic is that which can not be explained. You have now redefined what you mean by "magic" as "that which inspires wonder". Changing definitions after a series of comments is a pretty poor way to have a discussion.
- esafak 4y agoThat which can not be explained inspires wonder. That's basically what makes magic magic.
- gyrovagueGeist 4y ago"It doesn't stop being magic just because you know how it works" - Terry Prachett
- throwaway675309 4y agoJust because we have a pithy quote doesn't detract from the fact that we shouldn't go all starry eyed and bandy about terms which obfuscate rather than explore the mechanics of how things work. Calling something "black magic" is too close to all of the anthropomorphism bullshit that seems to travel like a miasma around LLMs.
- burnished 4y agoMagic is a garbage explanation for the function of a system, but the original context was some one expressing dismay at people being impressed and 'the thing resembles magic' is a great explanation for that.
- kerkeslager 4y agoSetting aside the silliness of that definition of magic, there's a huge leap between "I can't explain it" and "It can't be explained". There are plenty of explanations of how LLMs work, by their creators, incidentally.
- celestialcheese 4y agoYet there are emergent behaviours from these LLMs that are both surprising and not immediately understood. [1][2][3] Everyone has theories, of course, but still pretty "magic" considering these behaviours weren't theorised in papers prior to observation. 1 - https://www.jasonwei.net/blog/emergence https://www.jasonwei.net/blog/emergence 2 - https://arxiv.org/pdf/2206.07682.pdf https://arxiv.org/pdf/2206.07682.pdf 3 - https://www.quantamagazine.org/the-unpredictable-abilities-emerging-from-large-ai-models-20230316/ https://www.quantamagazine.org/the-unpredictable-abilities-e...
- kerkeslager 4y agoDon't cite stuff you didn't read or understand. [1] Is a summary of [2], by one of its authors, not a separate source. [2] Defines "emergent behaviors" in a way that you're clearly misunderstanding (because "emergent behaviors" is an extraordinarily poor way of communicating this--it's partly the fault of the researchers who chose this ambiguous language). All it's saying is that bigger models can do things that smaller models can't, which should be surprising to no one. It's NOT saying that the capabilities are anything more than the sum of the input data. [3] Is written by a journalist, not an AI researcher, and so it's limited by the things the journalist is excited about. The journalist, for example, downplays sections like, "The other, less sensational possibility, she said, is that what appears to be emergent may instead be the culmination of an internal, statistics-driven process that works through chain-of-thought-type reasoning. Large LLMs may simply be learning heuristics that are out of reach for those with fewer parameters or lower-quality data." If you're going to try to gather things from journalists rather than subject matter experts, you need to understand how journalists work, and how subject matter experts work, and look for paragraphs like that to understand what's actually happening.