10 week EMA is more logically powerful than a 10 day EMA. The reason being in weekly TF, 10EMA represents 50EMA in daily while 10 day EMA corresponds to much less important 50EMA in 75 min TF.

In the same way 40 week EMA is proxy to all important 200EMA in daily.

Hey, for people interested in this stuff, here's a thread about a change we made to running our facebook account:

So, about 5 years ago, we got on facebook. Why? Well, facebook has a big audience and they like to consume on facebook. So, we started posting comics there, even though we got paid nothing and facebook made money off the free content.

Why? Because back then, once in a while, you would say "also hey, I sell books" or "hey, I'm going to be signing at this event" and your facebook followers would actually see it.

Effectively, the implicit old arrangement was "facebook lets me reach readers efficiently, and I supply facebook with free content they run ads on."

Over time, like a bad business partner, facebook basically made it impossible to reach your audience without *paying them*. Thus, you supply them free content on which they run ads, and you get nothing in exchange.

Pretending that voting for Trump or denying climate change or suggesting the solution to every problem is more guns is a legitimate point of view, worthy of respect is not "how America works." It is how it dies.

People in a free society are entitled to any point of view no matter how obviously wrong or outlandish or destructive provided they do not impose their views on others.

But suggesting that something ignorant or divorced from reality warrants the same kind of treatment in public debate as something based in fact or at least credible is absurd and when done on behalf of a society in public media or academic settings it is self-destructive.

We would not grant media coverage or much bandwidth at all to a group that argued that unicorns exist or that there was a pot of gold at the end of the rainbow. Yet views like denying climate change or basing policy decisions on sweeping, obvious lies, are just as unsound.

Somewhere, somehow, judgments must be made. Some behaviors are wrong and must be condemned. Someone must challenge lies and demand facts. Coddling people who embrace idiocies does not help them no matter how loudly they demand to be coddled.

Important paper from Google on large batch optimization. They do impressively careful experiments measuring # iterations needed to achieve target validation error at various batch sizes. The main "surprise" is the lack of surprises. [thread]

https://t.co/7QIx5CFdfJ

The paper is a good example of lots of elements of good experimental design. They validate their metric by showing lots of variants give consistent results. They tune hyperparamters separately for each condition, check that optimum isn't at the endpoints, and measure sensitivity.

They have separate experiments where the hold fixed # iterations and # epochs, which (as they explain) measure very different things. They avoid confounds, such as batch norm's artificial dependence between batch size and regularization strength.

When the experiments are done carefully enough, the results are remarkably consistent between different datasets and architectures. Qualitatively, MNIST behaves just like ImageNet.

Importantly, they don't find any evidence for a "sharp/flat optima" effect whereby better optimization leads to worse final results. They have a good discussion of experimental artifacts/confounds in past papers where such effects were reported.