Audio note: this article contains 60 uses of latex notation, so the narration may be difficult to follow. There's a link to the original text in the episode description.
I think we may be able to prove that Bayesian learning on recurrent neural networks is equivalent to a bounded form of Solomonoff Induction, linking Singular Learning Theory (SLT) back to basic Algorithmic Information Theory (AIT). This post is my current early-stage sketch of the proof idea. Don't take it too seriously yet. I’m writing this out mostly to organise my own thoughts. I'd originally planned for it to be a shortform, but I think it ended up a bit too long for that.
Background:
I recently held a small talk presenting an idea for how and why deep learning generalises. Slides for the talk here, slide discussion here.
In the talk, I tried to reduce concepts from [...]
---
Outline:
(00:48) Background:
(02:47) Proof Outline:
(02:51) Setup: Predicting a stochastic process
(03:44) Claims I want to prove:
(08:47) Comments:
(10:30) Thank yous
The original text contained 3 footnotes which were omitted from this narration.
---
First published:
February 10th, 2025
Source:
https://www.lesswrong.com/posts/3ZBmKDpAJJahRM248/proof-idea-slt-to-ait
Narrated by TYPE III AUDIO.