So in @sirajraval's livestream yesterday he mentioned his 'recent neural qubit paper'. I've found that huge chunks of it are plagiarised from a paper by Nathan Killoran, Seth Lloyd, and co-authors. E.g., in the attached images, red is Siraj, green is original 
@sirajraval He cites Killoran et al. in the abstract only, and says he presents a model "similar to" it, but the shown parts are almost word for word the same, and the figure, table, and captions are also lifted
@sirajraval Some parts appear to have been copied with words replaced with nonsense 'synonyms'. E.g., in Killoran et al.: "There is a key distinction in the CV model between the quantum gates which are Gaussian and those which are not." becomes...
@sirajraval "In the CV model, there's a key difference between
Gaussian quantum doors and non-Gaussian ones" 'Gate' has become 'door', because without understanding on the part of the author of what they're writing, these are synonyms
Gaussian quantum doors and non-Gaussian ones" 'Gate' has become 'door', because without understanding on the part of the author of what they're writing, these are synonyms
@sirajraval More
@sirajraval And more. It was easy to find the plagiarism because even his equation numbers are the same: because he's just dumped in screenshots of the originals!
@sirajraval Sorry, I should provide links. Killoran et al.: arxiv.org/pdf/1806.06871…, Siraj: sirajraval.com/wp-content/upl…
@sirajraval Another. This part is sad, it's just a description of classical neural networks. @sirajraval is currently running an online course with over 1000 students charged $200 each in which he teaches them machine learning, and he has plagiarised this
@sirajraval By the way, I've chosen to expose this in a very public way because Siraj is currently running a course for over 1000 students at $200 a pop, and was recently accused of exploiting those students when he refused refunds. Now this. The man is a fraud and his students should know
@sirajraval Another nonsense synonym, presumably trying to hide plagiarism: "complicated Hilbert space"
@sirajraval Oops, I just realised that from my screenshots you can see that I've named my downloaded copy of Siraj's paper 'bullshit.pdf'
@sirajraval Hey @esa, you're due to have this guy run a workshop soon. You should probably rethink that cosmos.esa.int/web/esac-stats…
@sirajraval @esa In Siraj's github repo for this paper: "XanaduAI had a great proof of concept in this direction, I built off of their variational circuit code by modifying the fock basis truncation, adding an extra photonic quantum layer, and increasing the intial gate parameter seed"
@sirajraval @esa None of those apparently very minor changes to the model are actually mentioned in the paper, mind
Also, unless he doesn't mean what he writes, one of his changes from the original is... changing a seed?
@sirajraval @esa One thing I don't understand: this plagiarism is so brazen I don't know how he expected to get away with it. Is he counting on his fans forming a closed community that won't see or believe this kind of exposé?
@sirajraval @esa Someone has just pointed out to me that the example application of the model in his paper is fraud detection 🙂
@sirajraval @esa And of course that part is lifted straight from the original paper too
@sirajraval @esa 🚨 The parts of the paper that aren't plagiarised from Killoran et al. are taken from this paper by Wang et al. arxiv.org/abs/1905.12599 I've shaded the plagiarised parts of this page...
@sirajraval @esa This is bizarre: on this latest page he's even paraphrased the quotations!
I hate to have to tweet about this again, but people have reported to me that they can no longer see the first tweet in my thread about Siraj's plagiarism. Reported as spam/harassment? What can I do about this?
