Viewing profile — Mmrnmhrm

The power of FORTH is that creating a variation tailored to your specific problem is almost effortless. In most languages, building a compiler or interpreter is a major project. Wi…

I did. And when I went open with it, I did find out that plenty of people whom I know had breakdowns too. The hardest thing was to admit I needed help (used to be a high performer:…

>So HN, what has your experience been with depression? I am sharing my own experience, the topic being a taboo does not help people who suffer. I was also shit scared of medication…

I can relate. I just read this at 3am while trying to stop wondering what to do with my life. I had the childhood dream of being a professor, and after a short stint in industry, w…

I appreciate the answers. Thanks! I'm tying (unsuccessfully) to get sleep, and I'll try magnesium supplements too. I already spoke about this but never got solutions, and at this p…

If you have any resources on how to manage anxiety please bring them on. My old son has a mild development impairment, I am the sole bread earner, and we have no family support (we…

I understand this is a non linear control problem with a well known model. How would machine learning help there?

While the roots have been known for a long time, my impression is that the key paper that started this line of thought was Marco Cuturi's NIPS 2013 paper "Sinkhorn Distances", whic…

IMHO the coronavirus was used as an excuse to kill a fair that was already producing negative ROI. Due to the size and fame of the fair, most big companies were forced to participa…

We have double blind peer review precisely to avoid the bias that you express in your second sentence.

I'm not saying to discard this research. I'm suggesting to wait until it is peer-reviewed and published before jumping on it. To me, the capsule concept seems reasonable, and I hav…

Nice video, however instead of riding the hype train of arxiv, could we wait until peer review analyzes the paper? If someone other than Hinton presented a YADLA (Yet Another Deep …

I agree, it still needs a strong use case. IMHO, new network architectures are generally overhyped. Even AlexNet performs wonderfully well in most problems once you add nice initia…

I am very disappointed with constexpr in C++14 and C++17. The work shown here is impressive, but the point is that it should not be. D has shown that variable initialization at com…

http://www.doenercopter.de/ Similar, in Germany :)

that's a surprising move. Movidius had a rather unimpressive multiprocessor chip, but managed to get it incredibly hyped. IMHO, Intel is desperate to buy machine learning hype. Mov…

I dare Intel to provide an updated version of this paper, and I dare Nvidia to perform the peer review.