Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

I've been doing functional programming, primarily in OCaml, for a couple years now, and have recently ventured into the land of monads. I'm able to work them now, and understand how to use them, but I'm interested in understanding more about their mathematical foundations. These foundations are usually presented as coming from category theory. So we get explanations such as the following:

A monad is a monoid in the category of endofunctors.

Now, my goal (partially) is to understand what that means. Can anyone suggest a gentle introduction to category theory, particularly one aimed at programmers already familiar with a functional language such as ML or Haskell, with references for further reading? Resources not necessarily aimed at programmers but accessible to readers with a background in discrete math and first-order logic would be quite acceptable as well.

share|improve this question
1  
StackOverflow thread: stackoverflow.com/questions/1224491/… –  sdcvvc Dec 1 '09 at 23:07
add comment

8 Answers 8

up vote 18 down vote accepted

Online resources:

Books:

  • "Basic category theory for computer scientists" by Benjamin Pierce
  • MacLane - solid mathematical foundations, but hardly any references to computing
  • Abstract and concrete categories - might be considered too verbose, but it's full of examples

Category theory in Haskell:

Another list

share|improve this answer
    
I can recommend, strongly, the Barr/Wells above for application connection, and the Awodey book for the theory. And I'm glad to see my lecture notes listed! :-D –  Mikael Vejdemo-Johansson Oct 22 '09 at 22:56
    
I wikified the answer. Additions are welcome. –  sdcvvc Jan 25 '10 at 22:16
add comment

Since sdvccv already pointed out a number of good sources for learning category theory as applied to CS, I will try and provide some guide posts.

My favorite book on the subject is Practical Foundations of Mathematics by Paul Taylor since he does a really good job of giving you a big picture (unfortunately he doesn't always give you enough details if you don't already have a logic background). Bart Jacobs book Categorical Logic and Type Theory is also very readable.

In general I think the most important thing to understand in order to apply categories to computer science is the Curry-Howard-Lambek correspondence which loosely states that lambda calculii, intuitionist logics, and cartesian closed categories (categories where you have products and function spaces) are the same thing. Proofs and Types which was transcribed from some of Girard's lecture notes is an excellent source for the Curry-Howard part of the correspondence. Steve Awodey's book and these notes by Samson Abramsky are good places to see this translated into categorical language. For connections to Topoi Mac Lane's Sheaves in Geometry and Logic looks good.

Next you will probably want to learn about categorical and universal algebra. One of the more immediate and accessible applications of these ideas is the theory of algebraic data types (categorically: initial algebras for polynomial functors) and maps and folds between them. Monads are also a part of this subject since they are type constructors (endofunctors) that also have a multiplication and unit. Haskell do notation corresponds to forming the Kliesli category for a monad.

The nascent field of universal coalgebra has been very useful for formalizing notions of state and observation. There are also some emerging connections between coalgebra and modal logic.

Finally, if you aren't worn out you may want to learn about Stone duality which is a way of relating "logics of observable properties" and topology. For computer scientist Stone duality is primarily useful for giving a logical interpretation to domain theory, but mathematicians may recognize the duality between commutative rings and Zariski spectra as a special case.

share|improve this answer
    
Very good references. I think you can post using hyperlinks now. –  sdcvvc Oct 23 '09 at 14:09
    
Gah! Why can't I add references to this without bumping things to the first page! –  Justin Hilburn Jan 6 '11 at 18:14
add comment

Category Theory by Steve Awodey from Carnegie Mellon University has been a favorite text of mine. It's 266 pages, and while it's not a "practical theory book" (e.g., one aimed at programmers), it is very easy to follow and very well explained.

share|improve this answer
add comment

Categories for the working mathematician, by Maclane, is probably good for some working computer scientists too. No disputing taste, of course.

share|improve this answer
add comment

The book "Abelian Categories, an Introduction to the Theory of Functors" by Peter J. Freyd is a very good introduction to category theory. It is not aimed at computer scientists, but it is a good introduction to the subject.

share|improve this answer
add comment

TWF Week 89 for your specific question, and This Week's Finds in Mathematical Physics in general for wonderful presentations of all sorts of things category-related.

share|improve this answer
add comment

I'm skeptical of "practical category theory," especially books by mathematicians telling computer scientists how important category theory is. I'm not saying there aren't connections -- I know there are -- but there's a tremendous temptation for mathematicians to include gratuitous material.

share|improve this answer
1  
The Barr/Wells book is not very heavy on the gratuitious material - it puts quite some energy into the Sketch theory formalism (Wells' big thing), but otherwise does a good job of relating almost everything to various corners of computer science. –  Mikael Vejdemo-Johansson Oct 22 '09 at 23:00
add comment

You may find Computational Category Theory by Rydeheard and Burstall interesting. It's available online at Rydeheard's page: http://www.cs.manchester.ac.uk/~david/categories/

The Manual there is the textbook and The Programs provide a great deal of the ML code in the text in its development of category theory.

For someone new to CT it's very different than most introductions but the code helps make some of the ideas much more concrete.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.