HackerTrans
TopNewTrendsCommentsPastAskShowJobs

ekez

no profile record

Submissions

Fast Matrix Multiply on an Apple GPU

percisely.xyz
2 points·by ekez·9 mesi fa·0 comments

Gradient Trouble

percisely.xyz
1 points·by ekez·anno scorso·0 comments

Gödel's Completeness Theorem

percisely.xyz
3 points·by ekez·2 anni fa·0 comments

comments

ekez
·7 mesi fa·discuss
I implemented an algorithm for neural network verification called ⍺β-CROWN for a deep-learning library called tinygrad.

https://github.com/0xekez/tinyLIRPA

tinygrad’s small set of operations and laziness made it easy to implement. Tho my overall sense is that neural network verification is currently more of a research interest than something practical.
ekez
·anno scorso·discuss
I wonder if the authors have tried incorporating error feedback from Lean into their models.

Work from 2023 [1] showed general purpose models did better when they were able to incorporate error feedback, humans incorporate error feedback, but none of the SOTA models on minif2f seem to.

[1]: https://arxiv.org/abs/2310.04353
ekez
·anno scorso·discuss
There’s something compelling about these, especially w.r.t. their ability to generalize. But what is the vision here? What might these be able to do in the future? Or even philosophically speaking, what do these teach us about the world? We know a 1D cellular automata is Turing equivalent, so, at least from one perspective, NCA/these aren’t terribly suprising.
ekez
·2 anni fa·discuss
If "we" all got together and reduced permitting costs, "our" homes would be much more valuable because a developer can build a highrise where the house is.
ekez
·2 anni fa·discuss
Thank you for the response. (I assume you mean "p' < -eps" in the second conditional?)

How often are rational approximations computable within any given tolerance?
ekez
·2 anni fa·discuss
why is it impossible to decide whether x<0, x=0, or x>0 as in Example 1?
ekez
·2 anni fa·discuss
The metric the authors use confuses me.

Edit distance seems like a strange way to test if the model understands arithmetic ([1], Figure 3). I think `1+3=3` would be equally as correct as `1+1=9`?

Why not consider how far off the model is `abs(actual-expected)`? I wonder if there is an inflection point with that metric.

https://arxiv.org/abs/2206.07682