Freeman Dyson - Fermi's rejection of our work (94/157)

What Hacker News said

Recommends EdwardCoffin · 2024-07 · on “Fitting an elephant with four non-zero parameters”

Freeman Dyson recounts the episode [1] that inspired this paper in his web of life interviews (prepositioned to the fitting an elephant bit) [2] [1] [this video] [2] [this video]

Recommends Calavar · 2024-07 · on “Fitting an elephant with four non-zero parameters”

> > In 1953, Enrico Fermi criticized Dyson’s model by quoting Johnny von Neumann: “With four parameters I can fit an elephant, and with five I can make him wiggle his trunk.” For those who are interested, you can watch Freeman Dyson recount this conversation in his own words in an interview: [this video]

Cites magicalhippo · 2024-03 · on “Study: Dark matter does not exist and the universe is 27B years old”

Full paper here[1], and here's[2] the earlier paper about the model itself. They overloaded the CCC acronym, which has also been used for Penrose's Conformal cyclic cosmology[3], though can't say which is more speculative. I'm also reminded of Von Neumann's elephant[4], given that they vary the constants it doesn't seem that surprising I suppose that they can get a fit. Freeman Dyson learned this the hard way[5]. This is just one test, the article mentions a few others that they'll need to pass to be a serious contender. Will be interesting to see how it goes. [1]: [LINK] [2]: [LINK] [3]: [LINK] [4]: [LINK] [5]: [this video]

Cites kayo_20211030 · 2024-09 · on “What Is a Particle? (2020)”

I found the reference. It was an interview with Freeman Dyson where he said "quarks were invented", and I don't think it's a slip of the tongue. It's at 3:17 in this video of an interview with him [this video] It's representative of a view that there's a thing (a particle) that explains another thing (a force) that was consistent with both theory and experiment. Thus, a quark could be a particle, and it was whatever the experimentalists and theorists said it was, however it was measured or contemplated. For a deeper meaning it becomes an exercise in hermeneutics i.e what does it mean when we say "particle"? That was the point of the original piece - there is no uncontested view of a particle's fo

Cites godelski · 2024-11 · on “A statistical approach to model evaluations”

I haven't had a chance to read that, but that quote suggests I should (especially considering the author and the editors). I often refer to "Elephant Fitting" w.r.t these systems. I suspect you understand this, but I think most think it is just about overfitting. But the way problem isn't about the number of parameters, but that parameters need to be justified. As explained by Dyson here[0]. Vladimir's quote really reminds me of this. Fermi likewise was stressing the importance of theory. I think it is a profound quote, and you were (are?) lucky to have that friendship. I do think abstraction is at the heart of intelligence. François Chollet discusses it a lot, and he's far from alone. It se

Cites godelski · 2025-01 · on “The impact of competition and DeepSeek on Nvidia”

Interestingly a lot of the math and physics people in the ML community are considered "grumpy researchers." A joke apparent with this starter pack[0]. From my personal experience (undergrad physics, worked as engineer, came to CS & ML because I liked the math), there's a lot of pushback. - I've been told that the math doesn't matter/you don't need math. - I've heard very prominent researchers say "fuck theorists" - I've seen papers routinely rejected for improving training techniques with reviewers say "just tune a large model" - I see papers that show improvements when conditioning comparisons on compute restraints because "not enough datasets" or "but does it scale" (these questions can al

Cites godelski · 2025-02 · on “Understanding Reasoning LLMs”

I don't think this makes sense and I'm not quite sure why you went to ML, but that's okay. I am a machine learning researcher, but also frustrated with the state of machine learning, in part because, well... you can probably see how "proof by empirical evidence" is dialed up to 11. Sorry, long answer incoming. It is far from complete too but I think it will help build strong intuition around your questions. Will knowledge transfer? That entirely depends on the new problem. It also entirely depends on how related the problem is. But also, what information was used to solve the pre-transfer state. Take LLMs for example. There's lots of works that have shown them being difficult to train for so

Cites godelski · 2025-03 · on “ARC-AGI without pretraining”

> I feel like extensive pretraining goes against the spirit of generality. What do you mean by generality? Pretraining is fine. It is even fine for pursuit to AGI. Humans and every animal has "baked in" memory. You're born knowing how to breath and have latent fears (chickens and hawks). Generalization is the ability to learn on a subset of something and then adapt to the entire (or a much larger portion) of the superset. It's always been that way. Humans do this, right? You learn some addition, subtraction, multiplication, and division and then you can do novel problems you've never seen before that are extremely different. We are extremely general here because we've learned the causal rule