Viewing profile — cgadski

Does anyone understand how/why old HN accounts become mouthpieces for language models?

No marketing bots on HN please.

https://cgad.ski

I was a little informal with my argument. It's not strictly true that we only see C = 0.2 when K = 2. I was reading what the graph says about the case when N is much greater than k…

There's a lot of beautiful writing on these topics on the "pure math" side, but it's hard to figure out what results are important for deep learning and to put them in a form that …

> The implications of these geometric properties are staggering. Let's consider a simple way to estimate how many quasi-orthogonal vectors can fit in a k-dimensional space. If we d…

The technical report says (page 7): > Our architectural choices are closely aligned with principles observed in biological brains. How? They point out three design choices: linear …

It's not really about the language. If someone doesn't speak English well and wants to use a model to translate it, that's cool. What I'm picking up on is the dishonesty and vapidn…

> This blog post has explored the most critical equations in machine learning, from foundational probability and linear algebra to advanced concepts like diffusion and attention. W…

To summarize: we're making a series of i.i.d. draws from a distribution and asking how many draws N we need to make until we get something larger than our first draw. Conditional o…

It's server-side! I render the latex with katex as part of my static generation. See: https://katex.org/docs/api#server-side-rendering-or-renderin... .

In little bits of free time I get here and there, I've been working on using reinforcement learning to build some better bots for my favorite multiplayer game. Project is up here: …

Where x is the final hidden layer of the base model, the idea here is to steer outputs in some direction by adding a vector y. More specifically, y is an exponential moving average…

Yeah, I think it's really important to understand how to coax non-equivariant models into being equivariant when needed. I don't think purely equivariant architectures are the way …

One way to understand why without writing down the CDF/PDF: When X is an exponential variable and c is a constant, X + c has the same distribution as X after conditioning on large …

I'm very excited that we're figuring out how to use deep learning on small numbers of data points! I'm curious about the focus on information compression, though. The classical vie…

Hm, so maybe the following is one answer. In this article, we fix a mereology and a kind of quantity Q that "decomposes" over it---in the sense that Q(p) = sum_{r In other situatio…

Super incomplete thought: how does this point of view relate to Euler characteristic? Can I get to Euler characteristic by asking how to solve an equation for some quantities q in …