back
user profile
srean
8,389karma·3,978submissions·October 4, 2010
about
cat > Bourne.identity
#! /bin/bash
echo "$*"
^D
recent activity (3,978 total)
comment
> Do you think terrorists are really going to name their Bluetooth speaker "bomb"? Two comments. If they did and no one took any action people would be asking for their (authority's)…
comment
Quantile regression is such an under-used tool. It really needs to be better known.
comment
It might help to pose the problem as regression analysis with with categorical dependent variables. Perhaps with generalized linear model to account for the case that the predicted variable is positiv…
comment
Loved them. Wish they were more realistically animated :)
comment
You might like having a go at Lush. It has fallen out of favor of late but is a very interesting language/system. https://scottlocklin.wordpress.com/2024/11/19/lush…
comment
Sleep deprived, I interpreted it as Gun Russian splat. Makes for whole different meaning, not that I approve the altered meaning in any way.
comment
Such a beautifully done site. I might be in love already . Many kudos. I dropped in a suggestion to do one on an umbrella. There's a lot going on in these. One can study the differential geometry…
comment
Was entirely tongue in cheek. A long time admirer of your blog. Having a wonderful wife myself I can understand the feelings.
comment
To address the degree bit, one could bake a couple of pies once in a while. Let me show myself out.
comment
Europe needs to get it's shit together. It has very not been together.
comment
Wait, wasn't that post revolution USSR / Mao's China ? Or in their words, only correct science is "Marxist" science
comment
I wonder if it was ever considered to page size the first new page proportional to the current submission rate. The idea is that with a fixed page size a new submission would get evicted off the first…
comment
Good compression, necessarily, requires good "next token" probability estimation. I was essentially using zlib's implicit probability estimator.
comment
Paging sdenton4. Once when I was in grad school I noticed an announcement for a lighthearted programming competition for rock-paper-scissors playing agent. The deadline was barely hours away and I was…
comment
Absolutely love your line-scanned trains. Hope that is still going swell.
comment
SIG Bovik ?
comment
https://en.wikipedia.org/wiki/The_Mansions_of_the_Gods
comment
It is amazing how much of 'it' is going on in the skies once you start noticing. There was a thread this winter on solstices, with commenters expressing wonder at how could humans figure th…
comment
> Isn't purple expected to be there per ROYGBIV I do not understand the Physics of the phenomenon very well, but I would expect the high frequency wavelengths (for example, purple) to be scatt…
comment
What I find fascinating objects of beauty are the ways different clouds modify light or are illuminated by different kinds of light, from the belt of Venus to different kinds of halos to crepuscular a…
comment
Noted.
comment
This is true in the absolute, not just in this special case but in almost all national level elections of any democratic country. I am going out on a limb here, but I think this is mostly true. "…
comment
As an outsider to the US it feels as if Trump's popular base is still with him, or am I misreading the room ? Downvote patterns would be revealing. I suppose his base considers people like her as…
comment
This bit tangential to the thrust of the post but what I find so interesting about Simpson's paradox and statistical inference / machine learning in general is that the same techniques that …
comment
AI is too nebulous, too I'll defined to be used without quotes, almost serious. Also the 'prompt' is not a textual prompt of course, but its vector embedded representation.
comment
There are a few comments on the unnatural/hallucinated look of the compressed images. I don't like it, but this is sort of the expected behavior for 'AI' based compression and deno…
comment
This is a potential problem in "AI" denoising as well. These denoising models, the autoencoders more directly so, work by (lossily) mapping the raw input to a very low dimensional representa…
comment
Very interested in reading your list of blue bananas in Oppenheimer.