It's too bad this article never links to the paper [1] being hinted at. In late 2012 (first submission) tensor decomposition was likely very interesting (the p

It's too bad this article never links to the paper [1] being hinted at. In late 2012 (first submission) tensor decomposition was likely very interesting (the paper has >1000 citations!). But given that this article doesn't distinguish between "use tensors" and "use tensor decomposition, not SVD or similar for old school setups" I think the article is super confusing. Perhaps the author didn't get the distinction, and decided to write up a "renegade/rebel" piece that is sadly a non sequitur. [1] https://arxiv.org/abs/1210.7559 https://arxiv.org/abs/1210.7559