Zulip Chat Archive

Stream: computer science

Topic: Proof Assistant Questions


Ender (Oct 08 2025 at 16:08):

The BNF derived parse tree of a proof is like a deep structure in generative grammar, I think. Can I use generative grammars for proof search? Feel free to correct my misunderstanding. This would entail exploring the possibility space of proofs instead of proving one theorem in particular. 

  • The closest I can find to a proof search algorithm that uses rewriting or replacement rules would seem to be proof search algorithms using sequent calculus. 

  • Is Lambek calculus at all related to this concept? Lambek calculus and the Lambek grammar keep coming up in my searches because I am interested in generative grammars.

What are the building blocks of a proof assistant?  (Proof calculus, logic calculus?)

What is the difference between a proof calculus and logic calculus? 

What does it mean for a calculus to be uninterpreted?


Last updated: Dec 20 2025 at 21:32 UTC