Higher-order unification

Question

I\'m working on a higher-order theorem prover, of which unification seems to be the most difficult subproblem.

If Huet\'s algorithm is still considered state-of-the-

Answer 1

I would add to the reading list a chapter in volume 2 of the Handbook of Automated Reasoning. This chapter is probably more accessible to the beginner and ends with λΠ-calculus where Conal Elliott paper starts.

A preprint is found here:

Higher-Order Unification and Matching
Gilles Dowek, 2001
http://www.lsv.fr/~dowek/Publi/unification.ps

Conal Elliott paper is more formal and concerntrated on one variant, and also introduces a λΠΣ-calculus in the end, which has also sum-types besides product-types.

Bye