vermarish

9 ms·

1.
by vermarish 6d ago
For over a decade, Google's been wearing the environmentally-friendly / carbon-neutral label as part of their public-facing brand values. It's definitely been eating into their CapEx then, and this isn't much different.
2.
by vermarish 26d ago
Fair enough!
3.
by vermarish 27d ago
> "Click any region to stimulate it" > It's just a markdown-embedded png. What did I expect?
4.
by vermarish 1mo ago
The link now says it's extended til August 31st.
5.
by vermarish 1mo ago
This is beautiful work! First, I thought this would involve using PCA to go from 3D to 2D, which would result in an easier selector to use, but at the expense of representing every person. Then, I thought this would stop at using the U-spac
6.
by vermarish 1mo ago
When you have Claude Code indepedently author commits and PRs and merge them in, it'll always credit itself as an author. I assume it showing up in the README is a byproduct of the same logic.
7.
by vermarish 2mo ago
Every time I come across a pudding article, it stops me in my tracks. They're always so rewarding!
8.
by vermarish 2mo ago
Not me, but my dentist always keeps one of the Monterey Bay Aquarium's livestreams in her office. Sea otters: https://www.youtube.com/live/abbR-Ttd-cA Jellyfish: https://www.youtube.com/live/e
9.
by vermarish 7mo ago
I have the same issue. Perma-black loading screen that resolves as soon as I disable ublock origin. Using Firefox on Debian.
10.
by vermarish 1y ago
I interpret it a different way than that. I see application code and testing code as both a part of blue team. It's the code reviews and architectural critiques that are part of red team. Personally, I've found GitHub's featu
11.
by vermarish 1y ago
They made Claude Code available on their $20/month plan about two weeks ago. Your point still stands, of course.
12.
by vermarish 2y ago
I love the animation when you click "No" on "Do you want a new Pebble?". So extra.
13.
by vermarish 2y ago
(2023)
14.
by vermarish 2y ago
This is really cool! I've seen Jensen's inequality used many times over in my stats/ML classes, but the traffic example here gave me an "aha" moment about how it manifests. I like the visualizations of the expected
15.
by vermarish 2y ago
I'm not a huge fan of the color/pitch relationship they seem to be trying to establish. What I do appreciate is the engineering design in these interactives. The circular metronome in the rhythm apps is very cool, might be good fo
16.
by vermarish 2y ago
It sounds like Google wants to get more edge compute in people's homes so they have a new vector to deploy AI products on, but they're still so far from actually deploying an innovative product that they can't announce anythi
17.
by vermarish 2y ago
Very cool! Just curious, would you consider adding more exotic experimental design setups like Latin Square Design to the roadmap?
18.
by vermarish 2y ago
At a high level, Bayesian statistics and DL share the same objective of fitting parameters to models. In particular, variational inference is a family of techniques that makes these kinds of problems computationally tractable. It shows up
19.
by vermarish 2y ago
I feel like cost, availability, taste, and strength are kind of important actually.
20.
by vermarish 2y ago
I didn't expect Sweden's Prohibition history to be so interesting. https://en.wikipedia.org/wiki/Systembolaget#History
21.
by vermarish 2y ago
This seems really cool! I noticed the landing page has a rotating PNG gallery in desktop but not on mobile. I'm sure you must've wanted to put it in the mobile page too, was it hard to UX or just too inconvenient to implement easi
22.
by vermarish 2y ago
I don't know. I think the younger 20-somethings all have this same kind of dream, but the older people get, the more comfortable they seem hiding behind NDAs at FAANGs or staying in "stealth mode." They really don't owe
23.
by vermarish 3y ago
But for such a model, the joint pdf would not be written simply as a product of each individual pdf. That's what independence provides.
24.
by vermarish 3y ago
I mean, instead of using a fixed threshold at 500, if you use a live threshold determined by the recent average upvotes, then yeah I'd have no qualms with calling them outliers. It's just that this method is susceptible to votecou
25.
by vermarish 3y ago
Right. But if you make the notation slightly more explicit, then the integral of L(data, params) over data is 1. This follows from the independence assumption. So we ARE working with a probability function. Its output can be interpreted as
26.
by vermarish 3y ago
Hi! Some background first: I'm putting together a blog right now using Hugo and D3. I'm a huge fan of D3's infinite flexibility, as seen in some famous scrollytellers [0-1], and I've spent some time experimenting with th
27.
by vermarish 3y ago
Any explanation on the choice of the colormap? It can have a powerful effect on how we perceive this visualization.
28.
by vermarish 3y ago
Yes, this is a counterfactual line of reasoning. It's certainly legit, and I believe it's an active field of research in ML, but at the end of the day, you're still poking around the inputs of a black-box model as opposed to
29.
by vermarish 3y ago
As long as it spits out a p-value and the human decides it looks pretty small.
30.
by vermarish 3y ago
"Statistics" = models with interpretable decision outcomes "AI" = models without It's just a little bit easier to hold someone accountable for a logistic regression with predictive bias than for a neural network wit

More ›