My question concers the post-problem analysis techniques used to extract reusable patterns from problems, both solved and unsolved. I am considering journalling the complete thought process after each problem-recording how i interpreted constraints(eg n<=2e5 implying O(nlogn) as maximum time complexity), how i classified the problem(greedy,DP,binary search,ad-hoc), and what observations led to the final approach
I would like to ask experienced and high-rated problem solvers:what methods do you use to maximise pattern recognition and long-term retention during post-problem analysis? How do you ensure that insights transfer across problems rather than remain isolated?
It can be hard for me to explain my thinking process sometimes, so keep that in mind lol
Looking for patterns on hard problems will get you somewhere. If you know what an algorithm does, what its time complexity is, and its memory complexity, you'll definitely be able to use that knowledge when solving problems (which is quite obvious). But, trying to look for patterns in the problems won't help you solve every problem, or most harder problems IMO. Usually, I keep thinking of possible solutions that can become optimal solutions. Some of these ideas work, some don't. If you can, draw something. Don't think inside your head.
For "long-term retention," for me, I just kinda remember it. Bad answer to your question, but we don't necessarily try to remember a question for a long time. Again, not a great answer, but I can't really answer it differently.
Hope this helps!
Appreciate the response. I agree-exploring ideas helps more than forcing patters, and sometimes the useful insights tend to stick without trying to memorise them.