singhblom

It also has "Not yet" implications ...

All anglophone. I'm guessing privacy laws or something like that disqualifies the UK and Ireland.

But they have, right? Wegovy works really well. Ozempic is off label for weight loss only in the sense that it is the semaglutide drug that is marketed for diabetes and that it therefore isn't sold at the same high dose as Wegovy.

That's not what the article says that the court found though. It says that the Transport Agency needs to find another way of delivering the plates. The case was not against the postal service.

Dogs have more joints, one for the knee where the femur meets the tibia and one for the ankle where the tibia meets the tarsal bones. If you just have one single joint in the hind legs you probably want to be able to push off as easily as possible, so the lower one makes more sense.

You could require candidates to publicly disclose their ranking of the other candidates before the election. If they have to lock in their preference a few days before voting day that in itself would be informative for voters.

At least for the steel industry stuff quoted in OP the electricity consumption is mostly to produce hydrogen gas. The gas will then be stored in giant underground chambers and after that used in the steel production. This works extremely nicely with wind power since the electrolysis decouples the electricity consumption from the steel production. This means that the steel plants will act as stabilizers and modulators of the electricity prices in the whole region to a certain extent (together with the hydro plants and batteries presumably).

Aren't actor critic algorithms very close to GANs already? You have a generator/actor/policy that produces data and a discriminator/critic/q that says if the data is good or bad. The critic trains on the data generated by the actor and some extra info given by the user (rewards or example data) and the actor learns from the signal given by the critic.

I thought 花 and 鼻 were ambiguous in most romanizations - at least I would write it as "hana" in all romanizations I can think of, yet the pitch accent differs. No?

There was Russian influence, but maybe not for all the letters. Pinyin 'q' is based on Russian 'ч' if I recall correctly.

I don't think anyone would ever argue that a government should do things it shouldn't. The most common disagreement is not if the government should do more than it should, but what exactly it should do. :P

I don't know about this. My feeling from working with Typescript and Javascript is that the type hints really help people understand what is happening. My background is as a Python programmer, but Typescript really convinced me that optional type hints are a super-valuable addition to a dynamic language. Most of my docstrings in python tend to be type annotations anyway -- might as well make it possible for tools to support it easily.

They don't state flops because they don't do floats. The nervana chip has some weird fixed point format that they think works better for deep learning. I've heard similar noise out of Google (the other TPU company ...), so I wouldn't be surprised if we'll see the same in many of the specialized chips that people will build for deep learning.

I really don't think CUDA support is important for nervana's offering. They think of themselves as the Apples of deep learning - they want to offer an integrated stack from the chip all the way to the APIs. The way most people use deep learning you don't really need to know CUDA, you just need to use a library that is fast. So it's enough that nervana's engineers know how to write deep learning libs for their own chip. Furthermore I can't see Intel caring that much about CUDA support, since CUDA is owned by Nvidia.

What I've heard about the chip makes it sound really exciting. Many of the trade-offs in deep learning are different from the ones you do in graphics, so specialized hardware makes sense.