twtw

28 ms·

241.
by twtw 8y ago
The title is " introduction to mathematics." I think you are probably looking for a different book.
242.
by twtw 8y ago
I'm relatively unfamiliar with people who are actually arguing that deep learning is all that is needed for AGI - Andrew Ng can say what he wants, but do many of the researchers and engineers working with deep learning systems actually
243.
by twtw 8y ago
This could easily just be a lack of optimization for ARM. IIRC cloudflare reported a while ago that they were able to achieve 10x speed ups for some compression and cryptography tasks with relatively low effort.
244.
by twtw 8y ago
Thanks. I saw it was based on A72, but I couldn't tell if all the discussion of customization was referring to modifications to the core itself or just the SoC.
245.
by twtw 8y ago
Anyone know what kind of ARM license Amazon has? Just a processor license or a full architecture license? I'm curious if Amazon is doing the actual CPU design.
246.
by twtw 8y ago
Depending on how you define "properly written." It was easy to write seemingly proper code that assumed TSO and worked for x86 but does not work on architectures with weak memory consistency.
247.
by twtw 8y ago
> made the TPU and open sourced it In no sense is the TPU open source. If having an open source framework that is able to use the hardware doesn't mean it is open source - or if it does mean that to you, GPUs are open source too.
248.
by twtw 8y ago
> - which I believe is the case - then so is a naked goto. It's the cognitive dissonance of people who use them heavily while insisting that goto is absolutely evil, no exceptions, that irks me. Sounds like we are in agreement then.
249.
by twtw 8y ago
quote from previous post that is potentially relevant to your dislike of the "superior, disdainful tone": > later someone sent me a glorious photoshopped screenshot (see above) which shows me with a painted-on mustache and info
250.
by twtw 8y ago
Downsampling alone will not achieve this. It's typically done via phase vocoding or SOLA.
251.
by twtw 8y ago
To be honest, I have no idea what you are trying to say. Metacritic score is "highly, highly correlated" with film duration? This isn't science or engineering, it's art. You can't objectively determine whether any f
252.
by twtw 8y ago
Sabine Hossenfelder is a woman.
253.
by twtw 8y ago
> any reasonable objective metric Any true objective metric would rank films would likely be uncorrelated with reviews - stuff like frames per second, duration, resolution, etc. I'm of the opinion that talking about "objectiv
254.
by twtw 8y ago
Averaging subjective metrics doesn't turn them into an objective metric.
255.
by twtw 8y ago
> objectively one of their worst films I'm very intrigued by this - what kind of metrics does one use to objectively rank Pixar films?
256.
by twtw 8y ago
No, it's not more clean in the way those are. If you already have a loop, use break or continue. If you don't, it isn't "more clean" to introduce a pseudo-loop just so you can use break.
257.
by twtw 8y ago
The solution to this is an Error label after returning success and a "goto Error" in case of failure. The pseudo- loop plus break is not cleaner in any sense - its only purpose is to avoid typing "goto."
258.
by twtw 8y ago
Can't help but think of this: > The only real performance advantages the GeForce3 currently offers exist in three situations: 1) very high-resolutions, 2) with AA enabled or 3) in DX8 specific benchmarks. You should honestly not con
259.
by twtw 8y ago
There's more to exams than memorization too, and cheating because you don't like memorization is childish. If the exams really test only memorization, drop the class.
260.
by twtw 8y ago
Yes. SpaceX surely knows this, and reckons that the cost is worth full reusability with reduced refurbishment cost.
261.
by twtw 8y ago
Eh, probably some places (e.g. Uber) would love that kind of thing.
262.
by twtw 8y ago
If an individual's philosophy is that you can always google stuff so you don't need to know anything, I don't see why that person should waste time in school.
263.
by twtw 8y ago
> It’s also worth pointing out that all ideal circuit devices are non physical This is not a useful statement - in this framing, all of circuit theory is unphysical. Other ideal circuit elements could exist in the sense that their exis
264.
by twtw 8y ago
This is true, but it is essentially the same situation as using the CPU for processing.
265.
by twtw 8y ago
"The language for the 21st century should be business oriented, English-like and not dependent on native types. The most exciting thing, we already have the language exactly like this! What do you think it is?" Not to me, no. I th
266.
by twtw 8y ago
I don't think you can blame the people for reading all the words and deciding not to vote on the survey.
267.
by twtw 8y ago
Thanks for pointing this out. There really isn't any way to tell that the rando survey at the bottom is not actually a survey. I read what I thought was the whole article, but demurred from participating in the "survey."
268.
by twtw 8y ago
I would be extremely surprised if the motivation for putting bfloat16 in tensorflow was not the TPU. That first public commit was ~1.5 years before TPUv2 was announced at I/O, so it was almost certainly already in development.
269.
by twtw 8y ago
> many times over Where "many times" ~= 2.2, based on the revenue data in the article.
270.
The Borg Complex: A Primer(thefrailestthing.com)
4 points by twtw 8y ago|0 comments

More ›