>>14343638What Knuth book are we talking about?
>Concrete mathematicsthis is an amazing book. It starts off simple and looks simple, but there's a reason why his classes were typically full of late undergrads and early grads. It's full of problems that teach you how to solve hard classes of problems in combinatorics and analysis. It's less focused on developing problems used to develop structure (you won't find an analysis problem about compactness in the book) but more on solving really tricky problems with analysis and counting. Picrel is 3 problems from the asymptotics chapter.
>TAOCPthey're great as *references* after you've studied other algorithms texts. If you do algo research, it's probably illuminating to go read them, but don't worry yourself on going through all of it. TAOCP is a great read, but it's largely encyclopedic. I skipped volume 1 and studied a lot of volume 2 because I was interested in random number generators. On the other hand, a book that's just as rigorous but is probably better for the subject is this:
http://www.nrbook.com/devroye/It's an excellent read and freely available through the author's website. Still an authoritative book.
>Is there more focused on the matter books related to theoretical compsci?That's like asking if there are more focused books related to pure mathematics. There are so many topics that it's hard to answer your question. Besides, Knuth *is* a focused study into a subject of TCS: the classical study of algorithms (ie algorithms a la 1985, separated by a few popular topics). There have been a lot of advancements since then.
