abeppu

29 ms·

391.
by abeppu 2y ago
I'm torn between two reactions: - 7.5% sounds like a small improvement for a structural change introducing ML profile prediction - a ~250KB model making improvements in an area where humans have been working for a long time seems impre
392.
by abeppu 2y ago
> That's kind of the point of democratic institutions I think the problem is that American "democracy" is frequently undemocratic. If a clear majority of voters support a project or policy, it may still be blocked through
393.
by abeppu 2y ago
I've sometimes wondered whether arid parts of the American Southwest could, through the right kind of permaculture interventions, support more biological carbon sequestration, more plant life, drive development of richer soil etc. But
394.
by abeppu 2y ago
Isn't it just that there's too much money and too big an industry (with lobbyists and campaign contributions)? People are not happy with the system but they're less dedicated to pressuring politicians than companies who depen
395.
by abeppu 2y ago
1. I've checked multiple places and consistently "civilian" is defined in contrast to military / armed forces/ police, but not in contrast to government in general. 2. But even among non-governmental groups, this cl
396.
by abeppu 2y ago
But also, since it's a software-addressable issue, and because this process is likely such a headache, I wonder if in a couple of years they'll have a version that picks its headlight configuration based on your location. You&#x27
397.
by abeppu 2y ago
> Aftershock II is believed to be the world’s first civilian-built rocket to reach an altitude of 470,000 feet. Surely this first line, bolded and right below the byline, isn't correct for our normal understanding of "civilian&
398.
A tech charity imploded over a statement on Gaza(theguardian.com)
4 points by abeppu 2y ago|1 comments
399.
by abeppu 2y ago
Note: title shortened to meet length limits.
400.
by abeppu 2y ago
> It's possible to adopt “adults are just giant kids” as a lens through which to see other people. A work colleague is unreasonably angry at being left off an e-mail chain—but don’t playmates always hate being left out? An elderly r
401.
by abeppu 2y ago
Not the main topic, but the article is in part on how people respond to changing normative pressures so I think it's fair: I'm surprised at the low level of information in this article given that it's written by the original
402.
by abeppu 2y ago
I'm a little surprised there isn't a giant k-way tie for minerals that have only been found once. Like, I get that minerals arise from natural processes, and it's a big planet, so maybe every process should have a potential t
403.
by abeppu 2y ago
So the works Cindy displays are beautiful but is wood turning the way to "make the most of your burl"? I feel like there's a value that's related the surface area you can get out of a given burl. If you have a burl, are
404.
by abeppu 2y ago
I think it lines up with OR in the sense that x:(S+T) <==> x:S OR x:T If you know x:S you may find conclude x:(S+T) without also stipulating that x is not a T. It may be permitted (if pointless) to use S+T even in contexts where S is
405.
by abeppu 2y ago
But quotemstr seems interested in something else, not specific to writing. > Past empires, lost civilizations, and ancient artifacts have always fascinated us. > There was no glorious but crumbled past empire to inspire them. They fig
406.
by abeppu 2y ago
... were the Sumerians the first? The first what ? The article discusses the Uruk period, of maybe 4000 - 3000 BC. In the middle east, there had been millennia of settlement and agriculture by this point. Gobekli Tepe is apparently ~9000 B
407.
by abeppu 2y ago
And in particular, I'm always curious -- how much of what we know of cuneiform is just because using clay tablets made examples more likely to survive to the present day? You can kind of imagine there being a soft temporal horizon for
408.
by abeppu 2y ago
I don't know that efficiency is really an explanation. If you look at a list of very aerodynamic cars, there are a bunch of older ones with very different designs. https://carbuzz.com/features/most-aerodynamic-cars
409.
by abeppu 2y ago
I don't know who you are or what your qualifications are, but at least some of what you've said is in conflict with guidance I have seen from qualified professionals (some of which were in mutual disagreement). I'll note that
410.
by abeppu 2y ago
I'm not saying this doesn't happen, but I also think it's genuinely difficult to write policies that apply to technical systems that don't exist at the time of writing and which are also clear enough that regulators, cou
411.
by abeppu 2y ago
It's also just embarrassing that this is supposed to be a data science blog about employee performance and the only non-simulated data directly presented or discussed is the US wage distribution, where the author has just cavalierly ma
412.
by abeppu 2y ago
While it's great that this is open source, and I understand the pressure for smaller models that can be run in a wider range of contexts, I continue to be annoyed that authors keep posting comparisons to models which are slightly small
413.
by abeppu 2y ago
I'm fine with the claim that CS is not about computers as astronomy is not about telescopes, but there are pure CS courses that don't cover programming and do cover automata, turing machines, computability, complexity etc and don&
414.
by abeppu 2y ago
The algorithmic family that in general seems called for is "sequential monte carlo" or "particle filtering". While often this is presented for models and problems where the latent dimension is small and the problem is ab
415.
by abeppu 2y ago
> In fact, I’d go further and say that it’s the only computer science book of that age that I’d happily and usefully read again without it being just for historical interest: the content has barely aged at all. That’s not all that unusu
416.
by abeppu 2y ago
I can't tell if you're deliberately ignoring the point: people's public hobby projects may not be _about_ their area of expertise. They may be side projects specifically because they are outside of their main area. It isn&#x2
417.
by abeppu 2y ago
Even if an ML/AI/software engineer has a public GH with projects on it, there's no strong reason to expect it will be a useful signal about their expertise.
418.
by abeppu 2y ago
> The idea that "human-like" behaviour will lead to self-awareness is both unproven (it can't be proven until it happens) and impossible to disprove (like Russell's teapot). I think Searle's view was that: - whil
419.
by abeppu 2y ago
If you hire an accountant, do you expect to see the books of their other clients? When you choose a doctor, do you expect to see the charts of their prior patients? And frankly, when you hire a manager or executive, there's not general
420.
by abeppu 2y ago
While it's true that there are things that cannot be directly measured with data , that point cuts in both directions. Perhaps some rare and critical person who is happy in the RTO environment will create something of extraordinary val

More ›