[omeysalvi]

I really appreciate how writers at Quanta turn extremely complex and dry topics into a pleasurable read by mixing simple analogies with history. I really admire the skill it takes to break down these topics and make them fascinating for someone with no understanding of them, such as me.

[pmontra]

I like those articles too but I wish they include a short version with the major takeaways. In another industry that would be an executive summary. I don't always have the time to read all the story, so I end up fast reading it trying to find the important points and I'm never sure I really found them. In this case they seem to be the paragraphs after "Here’s an extremely rough cartoon version of the approach:"

[hammock]

Traditional calculus is not well-suited to describing the behavior of particles at extremely small scales because it leads to infinite values in calculations: Traditional calculus is based on real numbers, which are continuous and can take on any value. However, at the subatomic level, particles behave more like waves than like particles with a definite position, and the value of some physical quantities, like energy, can become infinite at certain points, which makes the equations used in traditional calculus break down.

To overcome the issues with traditional calculus, alien calculus involves replacing the standard real numbers used in calculus with "fuzzy numbers" that represent probabilities instead of definite values. These fuzzy numbers allow the equations to describe the behavior of particles more accurately because they capture the uncertainty inherent in subatomic systems.

Alien calculus calculates the probability that particles will be in certain positions or states, rather than trying to determine their exact positions or states. This is because particles at the subatomic level are constantly in motion and their position or state cannot be precisely determined at any given moment. By using probabilities, alien calculus can provide a more accurate description of the behavior of particles at this level.

[rolenthedeep]

Wait, is this new? This is exactly how I'd assumed quantum calculations worked already

[l33tman]

Yeah the above comment and it's reply is exactly what I'd assume GPT-4 would reply if asked what "alien calculus" is ;) (the linked article isn't about this at all, it's about technical details of perturbation series in mathematics as they relate to non-perturbative treatments)

[hammock]

You are correct that the probabilistic nature of quantum mechanics has been known for many years. However, what is new in the approach of "alien calculus" is the use of a mathematical framework that replaces traditional calculus with a more probabilistic approach to avoid infinities and better describe the behavior of particles at the subatomic level. Alien calculus is a novel approach that aims to improve upon traditional methods and has shown promise in resolving some of the issues that arise in particle physics calculations

[mathattack]

Thanks! Would you be aware of any kid-friendly sources that could be used to inspire students? Of the “We love calculus because it explains how a ball moves. Now let me show you something really cool!”

[hammock]

There is a nice discussion of quantum mechanics here - note you have to click to "reveal all steps" at the very bottom - https://mathigon.org/course/probability/randomness

Also check out https://ed.ted.com/search?qs=quantum+mechanics

[richk449]

Product opportunity: an LLM based tool to summarize works ti a personally customizable level. An executive will want the societal implications without the math. A scientist will want the equations without the analogies or number of elephants in size. A lay reader mag want to know the number of elephants without those equations getting in the way.

Could be combined with the information bubble filter described in Stephenson’s Burn or Dodge in Hell.

[knowsuchagency]

https://kagi.com/summarizer/index.html?url=https%3A%2F%2Fwww...

[Ar-Curunir]

Sometimes the fun is in the journey, not (only) the destination

[AnimalMuppet]

And sometimes it's not. Sometimes the journey is miserable, and we only do it to get to the destination.

And of the authors who think "my journey will be worth the destination, because I can really write!", some large percentage of them overestimate their ability and/or our interest level.

There was a time when people had more time to read than they had things to read. At that time, writing in a way that expanded the text made sense. But now, there's an essentially infinite amount to read, and still only 24 hours in a day. That favors getting to the point in a hurry. On the internet, you have from 5 to 30 seconds to persuade me that your thing is worth reading. If in that time you don't persuade me that you're going to deliver enough value to pay back my investment of reading your thing, I'm on to the next tab.

[Ar-Curunir]

That’s alright, not everything has to be catered to those with minimal time to read.

[AnimalMuppet]

It's not exactly minimal time to read. It's too many other options to read. If you look at the number of things that people (or at least websites) suggest I read, I've got an average of less than a minute to give each one (and maybe less than a second).

So if I'm going to read your thing, you have to pay me back more than those others are going to. You have to hold my attention more than the promise of all those other things.

But you're right that people different tastes. Some people like long-form writing. The same dynamic is still true in their choices, though - they're just using a different filter to decide what's going to pay them back the most.

[yuuuuyu]

All soon to be obsoleted by ChatGPT. /s

(I agree with your point and share your appreciation.)

[froggychairs]

Sarcasm aside. Even the output from GPT-4 is quite bland and generic. Very good for performing tasks (I.e convert a blob of text into a knowledge graph or generate X code), but quite awful at the elegant prose we see at play

Wonder how long before that’s solved. Hard to believe training on such large swathes of the web doesn’t result a compressed generic representation of language within its weights.

[simonh]

It seems like there’s a tension. On the one hand generating the most likely sequence of tokens maximises the chance the response will make sense and be relevant. On the other hand it also guarantees you will get the most bland and unimaginative response.

[maroonblazer]

I've only explored ChatGPT in the most cursory way, but wouldn't this be a function of the prompt it's given?

E.g. "Summarize this idea in a way that uses novel comparisons to everyday phenomena. The target reader is someone who doesn't have domain knowledge of the field...etc."

Or something like that.

[MacsHeadroom]

Yes, you're correct. People who complain that chatGPT is bland don't realize that they have to specify a style to not get the average of all content.

It's just like if you ask an image AI for a "woman" you will get the average of all artists women and it will look very generic and bland. But with the right stylistic qualifiers in your prompt you will get something so captivating that it wins awards for its creativity.

[froggychairs]

Prompting works very well, but I have still not seen any output from ChatGPT that captivated me in a way my favorite writing has.

I used to work in the field of creative text generation for fiction. I’m genuinely very curious and I go out of my way to find compelling examples. There is definitely “good” output that’s on the right track. GPT-4 also does way better, but it still falls short.

This is also difficult to just evaluate objectively. If you find that GPT models have generated the best prose you’ve seen. That’s wonderful! I understand that my standards are quite high (high does not equate to “better” either)

[RugnirViking]

that's exactly why it doesnt generate the most likely sequence of tokens! They are chosen at random based on the probabilities assigned by the model, so there is a chance of unusual output. In the API you can tweak the "temprature" which weights this towards more novel output

[froggychairs]

I’m very familiar with temperature and other parameters you can use to tweak output. They can take you decently far! GPT-2 can produce very coherent convincing output even today if you know what to tweak

Decoding methods also matter, and it’s a shame we aren’t given token probabilities (or any insight into model output) so we have more creative control over how to decode the output. Some of the better literature I’ve seen involving creative writing did have novel decoding methods

[GrinningFool]

Counterargument: "explain linked lists in the style of edgar allan poe"

[froggychairs]

My personal favorite is in the style of Yoda ;)

A tangent. Bard also confuses Yoda with Jar Jar Binks quite often. Try it out!

[serf]

>Wonder how long before that’s solved.

creative prompting gets most of the way there, already.

'Tell me the current weather as if you were Marcel Proust'

[froggychairs]

I replied to another comment with my thoughts on prompting. I will add that I don’t consider mimicking another writers style to matter much. Seems like an easy cop out (just my opinion though).

That’s not to say it isn’t impressive. It is and it accomplishes the job very well. We’re looking for progress not perfection, but I personally have very high standards from creative writing and GPT doesn’t meet my personal bar. However not everyone shares that bar and personal evals of GPT’s output are equally valid. Plenty of people find it to be great and at the end of the day that’s all that matters

[MichaelBurge]

Nobody outside OpenAI/Microsoft has access to the GPT-4 base model yet. Use Bing Creative for a differently finetuned GPT-4 than ChatGPT or "gpt-4".

[froggychairs]

GPT-4 comment isn’t true

Regarding Bing Creative. It is delightful! I do like it a bit and whatever they’ve done to the system does make for some of the better output I’ve seen from LLMs

[adamsmith143]

Seems mostly solved with prompting. I.e. "Rewrite this essay in the style of John Steinbeck." Works quite well.