Geoffrey Hinton's recent lecture at the Royal Institute[1] is a fascinating watch. His assertion that human use of language being exactly analogous to neural networks with back-propagation really made me think about what LLMs might be able to do, and indeed, what happens in me when I "think". A common objection to LLM "intelligence" is that "they don't know anything". But in turn... what do biological intelligences "know"?
For example, I "know" how to do things like write constructs that make complex collections of programmable switches behave in certain ways, but what do I really "understand"?
I've been "taught" things about quantum mechanics, electrons, semiconductors, transistors, integrated circuits, instruction sets, symbolic logic, state machines, assembly, compilers, high-level-languages, code modules, editors and formatting. I've "learned" more along the way by trial and error. But have I in effect ended up with anything other than an internalised store of concepts and interconnections? (c.f. features and weights).
Richard Sutton takes a different view in an interview with Dwarkesh Patel[2] and asserts that "learning" must include goals and reward functions but his argument seemed less concrete and possibly just a semantic re-labelling.
The vast majority of human learning is in constructing a useful model of the external world. This allows you to predict extremely accurate the results of your own actions. To that end, every single human knows a huge amount.
> learning when I can be doing something more efficiently
hardtime.nvim[1] (or vim-hardtime[2] if you're old-school) do exactly this but within your editing session. There's an associated blog post[3] explaining the rationale behind some of the workflow choices and you can of course bring your own.
I could swear the sky was reddish as I went to bed last night, in Stockholm... it was rainy, but as I turned off all lights I noticed an unfamiliar, slight red tint. I thought that maybe it was some neighbour's light reflecting (probably was) but seeing the Tasmanian skies in the photos made me think it might've been Aurora as the color is very similar!
I haven't seen anything in the news, but apparently some cities in Russia also reported Aurora sightings pretty far south.
Aunty sent me pictures of this night aurora in Latvia... it is more southwards than Stockholm. I was amazed. And sorry that I didn't know I had to be up to see it too for the first time in my life... eh
I was just racking my memory and searching through my library of interesting links to find exactly this!
Paul Chiusano gave a nice introductory talk at strangeloop last year: https://www.youtube.com/watch?v=gCWtkvDQ2ZI
Some recent posts have been talking about what happens to classical encryption in a post-quantum scenario. This article demonstrates a system capable of creating and distributing quantum keys at megabit-per-second rates.
[1]https://www.gregegan.net/PERMUTATION/Permutation.html