this reminds me of a book "monsters and magical sticks, there's no such thing as hypnosis?".
Which to my memory seemed to suggest hypnosis, and the states of altered consciousness associated with it, as people being affected by good storytelling. (don't think this was explicitly stated as such, but it the biggest takeaway I recall)
I suspect the social persuasion take here can be interpreted in a similar way; We are telling a story in concert with the people around us, and in that exploration of a narrative we assume roles that affect our perception. Or at least our own telling of this perception, which in turn may have effect on our retelling of it (both in terms of the rationalization we're doing internally, and in how we anchor the memories of what happened)
Going by that also, the knock-on is that regardless of whether it was "real" we will live our lives after based on our recollection of it, and assume in a sense that it was "real". Meaning that the results are more or less equivalent of it being real.
As a somewhat established researcher in the field, I second Jurasky and Martin. It is peerless and what I recommend to anyone joining my team if they think their background NLP knowledge is a bit on the weak side.
And as a lot of people have mentioned in here, DFT is pretty much implicated in neural networks already because of the mathematics (especially in convolutional/correlational neural networks, which often make use of the convolution theorem (which is "just" fourier coefficient multiplication) to do the convolution)
Extending this post it seems more interesting to look more generally at the correspondence with wavelet-transforms.
>especially in convolutional/correlation neural networks, which often make use of the convolution theorem to do the convolution
Is this true? With the learned filters being so much smaller than the input imagery/signals, and with "striding" operations and different boundary conditions being wrapped into these algorithms, it doesn't seem like a natural fit.
So... this isn’t news at all, but rather can be seen as supporting what our (norwegian here) own older tales and historians wrote down around the time:
https://en.m.wikipedia.org/wiki/Heimskringla
It’s well known we went around, even likely as far as asia (buddha statuettes were found in viking graves) and one of our kings (Harald Hårråde) was commander of the Byzantine royal Guard (https://en.m.wikipedia.org/wiki/Harald_Hardrada)
So... far as I see there’s no one who knows Norse history that would find this the least surprising.
Not really sure what the fuzz in the comments is about either. (Oh, and 13’th warrior was "based on" a, most likely, true event, with a, most likely, embellished narrative due to the fantastical language used back then to regale.)
The Russian Vikings (i.e. the Vikings who in the 9th century founded the original Russian state - Kievan Rus and accompanying dukedoms around - and ruled it for several centuries after that) were raiding as far as Caspian Sea and Iran and bringing the trophies - goods and slaves - for sale back as far as Scandinavia. So there were a lot of cross-continental blood/DNA mixing during those centuries (9th-12th) until Mongols came and that started another chapter.
Some years ago I watched an interesting documentary on how a bunch of viking swords that all bear the same distinctive engraving were very likely made of steel from Persia. The doc claimed that they were likely using a route along the Volga to get down to the Caspian routinely.
"Ibn Fadlan was a 10th-century Arab[1][2][a] Muslim traveler, famous for his account of his travels as a member of an embassy of the Abbasid Caliph of Baghdad to the king of the Volga Bulgars, known as his Risala ("account" or "journal").[b] His account is most notable for providing a detailed description of the Volga Vikings, including eyewitness accounts of life as part of a trade caravan and witnessing a ship burial.[4][5]"
"Initially, the Rus' appeared in Serkland in the 9th century traveling as merchants along the Volga trade route, selling furs, honey, and slaves. The first small-scale raids took place in the late 9th and early 10th century. The Rus' undertook the first large-scale expedition in 913; having arrived on 500 ships, they pillaged in the Gorgan region, in the territory of present-day Iran, and more to the west, in Gilan and Mazandaran, taking slaves and goods. On their return, the northern raiders were attacked and defeated by the Khazars in the Volga Delta, and those who escaped were killed by the local tribes on the middle Volga.
During their next expedition in 943, the Rus' captured Bardha'a, the capital of Arran, in the modern-day Republic of Azerbaijan. The Rus' stayed there for several months, killing many inhabitants of the city and amassing substantial plunder. It was only an outbreak of dysentery among the Rus' that forced them to depart with their spoils. Sviatoslav, prince of Kiev, commanded the next attack, which destroyed the Khazar state in 965. Sviatoslav's campaign established the Rus's hold on the north-south trade routes, helping to alter the demographics of the region. Raids continued through the time period with the last Scandinavian attempt to reestablish the route to the Caspian Sea taking place in 1041 by Ingvar the Far-Travelled. "
I would insists on differentiating Rus and Russia.
Rus was not the Russian state. If you want to associate it with any modern nation, then it would be Ukrainian.
But really, it was Rus.
P.S.
Just like ancient Roman state was not Italian state or Romanian state.
Fadlan's descriptions don't actually seem so fantastical & embellished to me, maybe he was that rare thing, a historian in the current meaning of the word.
Same difference back then.
(Iceland was basically expat Norwegians who held more on to the older beliefs while Norway got ravaged by christianity, and as a consequence of the priests going straight from dying rites to baptisms; the black plague.)
This reminded me;
People seem way too comfortable / in the dark when it comes to voice-cloning, which, with the added data of general speech-patterns, is likely to become a major issue wrt political/legal issues.
Conceptually, even though it currently is a market-strategy where the benefits are as you point out. In the longer term this could be argued as an uncomfortable seat of power to whoever sits on the user-data. (without bringing up too much of that issue here... might be time to generate obfuscation of voice similarly to the obfuscation of google-searching which I've seen on here earlier=)
Tragedy of the commons, and the laziness of a false sense of security railings.
Basically; I agree with you, and I think it is because of the way people act in a society that seems structured.
People seem to think that if there is a crowd of people, then they are part of the crowd. If a crowd sees someone in trouble, then the people are generally found to be more reluctant to step up and offer help, compared to when they are alone (plenty of research seems to show this.)
I suspect this is a sort of commons tragedy, where we assume that a crowd must hold structure with “people who take care of that sort of thing”, or that someone who feels qualified will somehow step up.
From my own experience it also seems like people tend to relegate any issues to whosoever looks like they have any kind of uniform. (Even employees at shops, etc.)
Were it not the case, we can speculate that people who are forced to bear their own actions would be more inclined to act out their own moral and ethical leanings... rather than the implied relation that illegal = bad, and legal/not-illegal= good (or acceptable).
In a sense then, anarchy is when all that is legal encompasses everything that falls under a sort of Kantian categorical principle. While everything illegal is the universally opposite. (Basically where the extrinsic laws of society match the intrinsic moral values of all citizens.)
Much used in simplifying kernel operations and convolutions (and some other nifty tricks.)
Another useful idea is also that in the domain of the fourier transformation we have exponentials (Fourier series are some series of $ c_n e^inx$), and when multiplying exponentials we get $ e^ix * e^iy = e^i(x + y) $
Moreover this is usually coupled with the case where we integrate on some periodic signal (so it’s integrated from 0 to 2pi, and unless the product of e^i(x+y) = e^0 = 1, then the integral becomes 0 as well. )
Which to my memory seemed to suggest hypnosis, and the states of altered consciousness associated with it, as people being affected by good storytelling. (don't think this was explicitly stated as such, but it the biggest takeaway I recall)
I suspect the social persuasion take here can be interpreted in a similar way; We are telling a story in concert with the people around us, and in that exploration of a narrative we assume roles that affect our perception. Or at least our own telling of this perception, which in turn may have effect on our retelling of it (both in terms of the rationalization we're doing internally, and in how we anchor the memories of what happened)
Going by that also, the knock-on is that regardless of whether it was "real" we will live our lives after based on our recollection of it, and assume in a sense that it was "real". Meaning that the results are more or less equivalent of it being real.