@eryceriousmatherfunker Very first step is to get a better idea of what specific subtopics are relevant to the position. You could maybe use this as a list of topics that might be relevant to electrical engineering.
Next week I'm gonna be interviewing an undergrad for a research position. (This isn't exactly a job, but kinda is because they'd be receiving course credit.) This particular person has an appropriate major (they're a neuro major and this is a neuro lab) but they lack coding experience though, which is an issue when coding is 95% of the actual work. So that brings me to my question: how can one judge innate coding talent / self-learning potential in an interview?
"Recently [experiment conducted in 2012, article written in 2015], by combining the highly sensitive receiving capabilities of the National Science Foundation’s (NSF) Green Bank Telescope (GBT) and the powerful radar transmitter at the NSF’s Arecibo Observatory, astronomers were able to make remarkably detailed images of the surface of this planet without ever leaving Earth."
but that's not the best that can be done... I think it would be more ideal for the pitch to jump around discretely and randomly. And to have the ability to focus on specific pitches while excluding others.
Tonight I'm going out on a limb and trying to listen to a slowly sweeping tone (pure sine wave going from 2500 Hz to 15000 Hz and back over the course of a couple minutes). My idea for this is mainly that I want to have the opposite effect of noise: to try my best to narrow the receptive fields of the neurons.
But regarding issue (2) there's also decent evidence that listening to broad spectrum unstructured noise can worsen overall network balance (as in, further reduce the strength of inhibition). This is also a result of "neurons that fire together wire together" because in white noise neurons across all frequencies are firing simultaneously.
Regarding issue (1) there's decent evidence that hearing a given pitch causes more neurons to become responsive to that pitch, giving rise to "notched sound therapy" in which a person listens to sounds/noise/music with the tinnitus frequency cut out, in order to nudge neurons that are receptive to the tinnitus frequency away from that cluster which has formed.
As in, in some fraction of people with hearing loss (I was in concert band for 12 years), two changes in auditory cortex occur: too many neurons get assigned to a single area of the frequency spectrum, and the overall level of circuit inhibition decreases.
Recently I've developed some noticeable tinnitus. I'm pretty sure I've always had some but recently it's gotten way worse. There's not really any proven effective treatment as far as I'm aware.
Actually my main comment is about the formatting of nash equilibria... for the RPS example up top, and the I/O section, you give the formatting as (strategy of A, strategy of B) but for the other test cases you use the formatting (strategy of A), (strategy of B)
In my case, it's not physically meaningful for the "particles" to leave the boundary of [0,1] so rescaling over time could normally be a smart way to accomplish the goal but it doesn't work in my case.
What exactly do you mean by rescaling? I think there's a way to map x -> x' such that either D(x') or u(x') is whatever function you want (like a constant).
But I don't know if that matters because there's ways to force the density function to be zero at the edges by having an infinite drift strength or diffusion coefficient at the edges.
the question is... does there exist some magical D(x) and u(x) functions so that, if p(x,t) is any Beta distribution, the future evolution will also always be a beta distribution?
consider a situation where you have 1D diffusion with drift. The substance starts out with a distribution of p(x, t0) and evolves according to the Fokker-Planck equation:
@El'endiaStarman I guess some more context: this has to do with IRL experience of cutting across rectangular fields of grass vs walking on the sidewalk around the edge. The simpler rule of thumb is "about half the length of the shortest side" because it takes 0.5 seconds to look and visualize your distance savings.
math approximation problem: take a right triangle. Is there a simple geometric construction that can approximate the difference between the hypotenuse length and the sum of length of the two legs (AKA, the amount of distance saved by "cutting the corner")?
It'd be a really nice thing to have though. There's already plenty of fake images floating around, only gonna get worse with DeepFakes. So it'd be really useful if someone could make a browser extension or something that's really easy for the public to use that identifies fake images as they are encountered.
There'd have to be some sort of web of trust among equipment manufacturers or something. The image is signed by the device which is signed by the manufacturer. But there'd have to be a registration process. I don't really think it's possible.
@DJMcMayhem could be either, I suppose. But it has to depend on image content, kinda like a hash, where there's no way to modify the image and preserve it.
Does there exist a cryptographic technique that could be used to prevent image manipulation? Or more precisely, something that could be embedded in pixel data that (1) links the image to the particular camera it was taken with, and (2) indicates no manipulation has occurred since the image left the camera?
