8000 GitHub - zhuoweiz/RL_for_ATP: Some code demo for setting up an reinforcement learning based proof search algorithm for classic FOL closed connection tableau
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Some code demo for setting up an reinforcement learning based proof search algorithm for classic FOL closed connection tableau

Notifications You must be signed in to change notification settings

zhuoweiz/RL_for_ATP

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Code demonstration for the paper presentation

A quick example using prolog

  • Install swi-prolog (lates version works).
  • run swipl to launch the swi prolog environment.
  • consult the example prolog file
?- [prolog_demo].

Result should show true.

A quick proof using LeanCoP

  • Install swi-prolog (lates version works).
  • Download and unzip leancop_swi.tar.gz 2.1 (http://www.leancop.de/leancopS.html#leancop10).
  • Go to the unzipped leancop folder.
  • run swipl to launch the swi prolog environment.
  • consult the leancop prolog file
?- [leancop21_swi].
  • Run a proof (if p is true, and for all X, p implies q(X), then for all Y, q(Y) is true)
?- P = ((p,all X:(p=>q(X)))=>all Y:q(Y)), prove(P, Proof ).
  • output should look like:
Proof = [[#, q(1^[])], [[-q(1^[]), p], [[-p]]]].

Which looks like this:

   q
 /   \
!q    p
      /
    !p

About

Some code demo for setting up an reinforcement learning based proof search algorithm for classic FOL closed connection tableau

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published
0