helltone

3 ms·

1.
by helltone 2mo ago
How do you think of uv vs pixi? Follow up: has anyone suggested the projects collaborate or eventually converge?
2.
by helltone 3mo ago
I thought glass was a solid?
3.
by helltone 9mo ago
I'm building in robotics. Setting up a new 3d camera today. I found that the 10m active USB C cable that I bought transfers power in both directions, but only transfers data in one direction, it turns out to be some weird video USB var
4.
by helltone 9mo ago
Is there anything like a linter to force you to stay within some subset of c++? I like the language, but it is hard to avoid language constructs that are outdated or enforce a single (or a few) ways of doing things. A c++ subset could be ni
5.
by helltone 11mo ago
In the UK, probably nuclear.
6.
by helltone 1y ago
The goal of automation is not necessarily reduce personnel, it could be increased reliability, reduced risk to life, reducing waste from scraps, faster lead time, etc
7.
by helltone 1y ago
Do you have some references for the pick and place and other reconfiguration things you mentioned. I've been out of this space for a while but last I checked these were still incredibly challenging things to get right.
8.
Ask HN: Tool that interviews engineers to capture project knowledge
4 points by helltone 1y ago|2 comments
9.
by helltone 1y ago
No amount of fine-tuning can prevent models from doing anything. All it can do is reduce the likelihood of exploits happening, while also increasing the surprise factor when they inevitably do. This is a fundamental limitation.
10.
by helltone 1y ago
Right I think I see it. This is insanely cool. But then there are performance tradeoffs in reusing intermediates vs recomputing that I think you can't represent. Some of these may affect numerical stability btw. See eg https:/&#x
11.
by helltone 1y ago
I have a background in program analysis, but I'm less familiar with the kind of kernels you are optimising. - Can you give some more insight on why 12 ops suffice for representing your input program? - With such a small number of ops,
12.
by helltone 1y ago
This heavily depends on share classes and preferences. Surely the new investor wants better terms. The issue isn't so much dilution as a preference but added risk of never even getting a payout at all.
13.
by helltone 1y ago
This resonates with me a lot. Can I contact you? My email is in my profile.
14.
by helltone 1y ago
Every time I see these headlines, the tech seems to be at least 10 years away from product. - demos done in a lab controlled environment without the crazy things that happen in a real world. - no humans nearby so none of the safety features
15.
by helltone 1y ago
This is very cool. I'm wondering if some of the templates and switch statements would be nicer if there was an intermediate representation and a compiler-like architecture. I'm also curious about how this compares to something lik
16.
by helltone 1y ago
No? Or maybe I'm missing something. If the goal is to be able to bound the computation of f, you can: 1) compute f with interval arithmetic 2) compute f normally and f' with interval arithmetic 3) compute f rounding towards zero,
17.
by helltone 1y ago
I think perhaps this could be done in other ways that don't require interval arithmetic for autodiff, only that the gradient is conservatively computed, in other words carrying the numerical error from f into f'
18.
by helltone 1y ago
How does the chart generation work under the hood? It's quite magical. Also how did you build the spreadsheet interface it's very cool.
19.
by helltone 1y ago
I'm building a tool to make ml on tabular data (forecasting, imputation, etc) easier and more accessible. The goal is to go from zero to a basic working model in minutes, even if the initial model is not perfect, and then iteratively i
20.
Nutpie: High-Performance Bayesian Inference(pymc-devs.github.io)
5 points by helltone 1y ago|0 comments
21.
by helltone 1y ago
Backtracking idea is interesting, could maybe diffusion help? At some point it turns into sat solving.
22.
by helltone 1y ago
Can I reach out to you directly to hear more? My email is on my profile.
23.
by helltone 1y ago
How do you modify ASTs?
24.
by helltone 2y ago
Program equivalence is undecidable, in general, but also in practice (in my experience) most interesting cases quickly escalate to require an unreasonable amount of compute. Personally, I think it is easier to produce correct-by-constructio
25.
by helltone 2y ago
Off topic but how is MCTS usually implemented efficiently? It has a branching structure that doesn't seem parallelizable (GPU).
26.
by helltone 2y ago
It seems GNNs operate on a fixed topology. What if I want to approximate some transformation of the topology of the graph? For example learning how to layout a graph, or converting program abstract syntax trees to data flow graphs.
27.
by helltone 2y ago
I'm curious and it's not obvious to me: what changed in terms of weight initialisation?
28.
by helltone 2y ago
None of the example tasks come with example solutions. Would you mind sharing what the solutions found look like?
29.
by helltone 2y ago
In addition to what is mentioned by the other comment, also threading and io.
30.
by helltone 2y ago
Off topic: I think there's some opportunities for making bayesian inference technology more accessible, and I'd love to chat with other people in this space. Email in my profile.

More ›