My biggest challenge with CS papers is figuring out how to actually read them. I've tried to read multiple CS papers and I still can't understand them. Am I doing something wrong?
1) Have a reasonable math background, or at least be willing to take it slow and google.
2) Read the abstract, intro, and conclusion and see if the problem and result are worth the effort.
3) Read the body. The whole body may or may not be relevant to you depending on your interest. If you are starting out in a field, try to read and understand the whole thing. You will learn lingo, what is expected of a published paper in the field, improve your math skills, and make reading other papers easier.
If you are having a lot of trouble with 1 or 3, find a MOOC relevant to the subfield and go through it then go back to the literature.
Read the abstract and confirm it is relevant to you.
In general, you will need to read papers if you want to be on the bleeding edge of the latest in research. That is machine learning research, and that is very specialized.
But you do not necessarily need to be on the bleeding edge to be an effective machine learning engineer. For that you can use often less abstract and more digestible sources, including MOOC courses, tutorials, books of multiple levels, vendor-specific docs, etc.
2) Read the abstract, intro, and conclusion and see if the problem and result are worth the effort.
3) Read the body. The whole body may or may not be relevant to you depending on your interest. If you are starting out in a field, try to read and understand the whole thing. You will learn lingo, what is expected of a published paper in the field, improve your math skills, and make reading other papers easier.
If you are having a lot of trouble with 1 or 3, find a MOOC relevant to the subfield and go through it then go back to the literature.