actually_a_dog

> Well that and it’s hard to distinguish if you’re early in exponential growth or linear.

That's true, but not really saying very much. Any differentiable function is locally linear around a neighborhood of any point where the derivative exists.

> Also exponential growth does hit some kind of ceiling relatively quickly.

Well... that depends. Much like how markets can remain irrational longer than you can remain solvent, exponential growth can often remain exponential for much longer than it takes to create a problem. Conversely, sometimes it can't remain exponential long enough to prevent a problem. Exponential growth is a hard beast to tame.

Yeah, unfortunately, I believe that one side effect of the human (sensory) nervous system having a logarithmic response curve is that humans are bad at understanding exponential processes. If you look at an exponential function on a log scale, you get a graph that looks linear. But, that's what fools people into thinking exponential processes are tame and easy to control.

Huge news, because this is the first improvement on the upper bound for the diagonal Ramsey numbers since 1935.

I am not. I am suggesting there are other savings in the average household energy budget that are more significant. Besides, even "waste" electricity heats a home in the winter, so it's not quite as bad as it sounds even.

20 watts standby is only about 14.7 kWh per month. Even at on-peak, summer PG&E rates, we're only talking about $7/month. Granted, I'd much rather pay $0 than $7, and I'd rather not waste energy, but we're probably not talking about anything close to the amount of energy the average household wastes. I'd be looking at refrigerators and other large appliances for energy savings long before I'd be thinking about how much electricity the TV uses on standby.

When this happened to me, I just renamed my tv to "STOP TRYING TO PAIR WITH THIS FUCKING TV"

By "diversity," do you mean something like "entropy?" Like maybe

    H_s(x) := -\sum_{x \in X_s} p(x) log(p(x))

where X_s := all s-grams from the training set? That seems like it would eventually become hard to impossible to actually compute. Even if you could what would it tell you?

Or, wait... are you referring to running such an analysis on the output of the model? Yeah, that might prove interesting....

As a former grad student, I can tell you, that's all research code, not just ML, or even "performance-oriented" research code.

Why? In my office, you'd see me (of course), and some bookcases behind me with a bunch of very SFW books. What's so offensive about that?

Don’t do that. Those blur effects always look so bad.