3 ms·
Ask HN: How to read deep learning/machine learning papers?
- billconan 9y agoI'd suggest searching for and reading blog posts of people who are studying deep learning. Academic papers are very pretentious, they are the tools for scholars to declare victories. They are not the best materials to understand. http://www.wildml.com/ http://www.wildml.com/ https://medium.com/mlreview https://medium.com/mlreview https://distill.pub/ https://distill.pub/
- ghosthamlet 9y agoThanks for you suggest. I just live math, and math + DL/ML/AI is very interesting, the DL framework/library is like normal software framework/library, easy and fun to use, but math relate papers are more fun.
- PaulHoule 9y agoWhen I read math-heavy papers I often read them over and over again over an extended period of time. Back in the early 00's I would get paper conferences proceedings out of the library at my local uni and read them on the bus, now I read on a tablet while I spin at the gym. If you really have to understand them, there is no substitute for working out the math partially or completely by hand, try to find a different way to derive things, reproduce the results computationally, etc. Do that on more than 1 or 2 papers then you will start finding errors!