Live data from Hacker News

Deep Learning with PyTorch: A 60 Minute Blitz [video]

pytorch.org

61–62 of 62 posts

Re: Deep Learning with PyTorch: A 60 Minute Blitz [video]

#61
post #38

Earlier quoted context omitted.

>Do you write your own crypto libraries too? Some people do. It's a good challenge.[0] [0] https://cryptopals.com/

To be fair, the main thing you learn doing the cryptopals challenge is to not write your own crypto. I had a lesson in writing crypto once, when I made what I thought was a good enough secret mixing procedure to encode some data I wanted to email outside of a company that didn’t allow web access. (Long time ago, circa 2000). It all looked undecipherable and I sent most of the data before I discovered that strings of…

>To be fair, the main thing you learn doing the cryptopals challenge is to not write your own crypto.

D'oh! Good point though lol.

Re: Deep Learning with PyTorch: A 60 Minute Blitz [video]

#62
post #21

One thing I've noticed is that it's quite hard to have vibrant discussions about DL because it is all either so simple or it is dauntingly complicated/unpredictable. Mostly my DL conversations end up being about frameworks. Anyone else experience this? Also the number of DL submissions on HN seems surprisingly low given the applicability of the technology.

It's pretty easy when you're talking to people who understand the fundamentals of deep learning, but that understanding isn't very common even on HN. I think that's because the real-world, valuable usecases of DL are not very accessible: (a) DL is pretty complicated in a way that's unfamiliar to most software engineers. You are consistently working with Tensors that have a couple more dimensions than people are used…

Could you list some of the places you look to learn? Where do you find relevant academic papers? It's hard to find information about what this learning workflow is.

Any recommended resources?

Post reply on HN