mirker

10 ms·

61.
by mirker 4y ago
There are other comments in this thread suggesting Altera shares were tanked due to manufacturing delays from Intel.
62.
by mirker 4y ago
That’s because recent focus on the space program is nonexistent. It’s been getting on life support since the moon walk.
63.
by mirker 4y ago
If you spend the time to contribute one patch, I don’t see why you couldn’t repeat this process to the point where you are highly productive. It’s just hard work getting to that point.
64.
by mirker 4y ago
One of the takeaways is some people enjoy pressure and some people use work to avoid pressure. You can find billionaires who have enough money to retire many times and they are still working 80 hour weeks. Why is that? They clearly enjoy th
65.
by mirker 4y ago
The downfall was fragmentation. They merged Keras after TF started having multiple competing application libraries, each managed by different people (and each fighting for promotion, as you say). Keras just happened to be the most popular.
66.
by mirker 4y ago
Ads companies getting rich off of AGI seems a bit sensational when they’re already getting rich off of the boring type of AI. They’ve already gotten rich indexing the web and all the data we have years ago.
67.
by mirker 4y ago
The book comparable to the mathematical level of CLRS, in my opinion. Which is either an intro book or very dense depending on a person’s math background.
68.
by mirker 4y ago
This type of complexity argument is also used in graphics rendering as far as I am aware. In graphics, it just has to look ok. Is there any concern for bounded error in MD?
69.
by mirker 4y ago
Yes, those are quite different. But you also can’t pipe lossy compression data willy-nilly without compounding data loss, so it’s hard to imagine an alternative with unix-style tools. The difference may also stem from those being applicatio
70.
by mirker 4y ago
As someone who learned about regular expressions before extensively using grep, I found grep to be quite unintuitive, since extended grep is what I am conceptually thinking about from a “theory” point if view. One difference between grep an
71.
by mirker 4y ago
Because there is an algorithm for it. Convert the strings into floating point numbers, add them, convert them back to strings. It’s a leetcode medium question. It should be learnable. The article talks about abstract math questions, but eve
72.
by mirker 4y ago
Ads companies like Google/Meta are good examples. Any small account gets a webpage and that’s it. Still, it must be harder to sell semi-commodities like CPUs.
73.
by mirker 4y ago
And then place persons in Faraday cage underground.
74.
by mirker 4y ago
It’s smoke and mirrors trying to fool you into thinking it’s generating intelligent text. In some applications e.g., a chatbot, that’s appropriate. But it’s really no comparison to an expert system for most applications, where you know exac
75.
by mirker 4y ago
Many of these scaling patterns are logarithmic with respect to data size. You can only double the dataset size so many times that it’s really not clear the scaling will continue.
76.
by mirker 4y ago
GPT3 can’t perform algebra over all 32 bit numbers. A trivial Python script can.
77.
by mirker 4y ago
It’s still more complicated because you have to amortize costs over several years and then also keep in mind that big players will design their own hardware to avoid the A100 cost and energy. Disks and other devices will also consume power.
78.
by mirker 4y ago
Yeah, I don’t believe the system is analog end-to-end. Moving the audio signal to every speaker across the entire length of the plane with analog seems like a fight against physics.
79.
by mirker 4y ago
Imagine building a compiler optimization pass. Do you want to hand-roll all the shifting equivalence relationships for division, etc?
80.
by mirker 4y ago
In hot fields, you have arXiv papers 2-3 generations ahead of peer review. Some have more citations than typical accepted papers. Peer review does not limit people’s attention.
81.
by mirker 4y ago
I agree scripting can be done top-level but it’d be shocking for it to be used for device management. These OS ideas are coming back to topics like microkernels, where you want to give applications part of the OS responsibility and hope the
82.
by mirker 4y ago
The dealbreaker is that consumer GPUs (and their drivers) have a terms of use prohibiting various aspects of datacenter usage.
83.
by mirker 4y ago
The norm is to use library functions like addition, multiply, power, etc., which have hardcoded derivatives. Then the derivative is computed “symbolically” by composing these base derivatives, which is essentially an exercise of implementin
84.
by mirker 4y ago
My nonstick explicitly says to cook on medium.
85.
by mirker 4y ago
Teflon vaporizes at >500F (and perhaps below). Leaving the pan on high without food in it will approach that quickly.
86.
by mirker 4y ago
You can only do teaching but it’s a different career path entirely. Some people don’t like it because you lose out on power and tenure.
87.
by mirker 4y ago
There isn’t an official safe exposure amount for lead [1]. [1] https://www.who.int/news-room/fact-sheets/detail/lead-poison...
88.
by mirker 4y ago
If the game was fixed and everyone had full information, there isn’t a reason this would happen. But I agree if the promotion/firing is relative or dynamic in some way, then you’d race down to arbitrarily low effort. The real problem i
89.
by mirker 4y ago
This seems cyclical to me. If you do the bare minimum and get fired, that is a contradiction. If the bare minimum is getting promoted then that means there is no room to slack. If you look like you are slacking to your coworkers, then you
90.
by mirker 4y ago
A rational actor will notice they get either a promotion or less work in the high/low work instances. If you work in between those boundaries, you get nothing extra.

More ›