I've uploaded my honours thesis, looking at the tactic semantics of the TWB.
I've uploaded my honours thesis, looking at the tactic semantics of the TWB.
I have ported the TWB to the new version of the ocaml compile. However since the TWB is heavily dependent on the old ocaml pre-processor, I've decided to postpone the re-write of the parser for the new pre-processor relesed with OCaml 3.10 and to use camlp5 instead.
Camlp5 is the old version of OCaml Pre-processor and it is now distributed as stand-alone library.
I've created a second branch of the TWB source code available via darcs .
http://twb.rsise.anu.edu.au/cgi-bin/darcsweb/darcsweb.cgi
Paper accepted to LPAR 07 : http://www.lpar.net/2007/
Abstract:
We are going to present this paper at a satellite workshop at TABLEAUX 2007 : http://tableaux2007.univ-cezanne.fr/scientific-program.html
paper submitted to AGENTS, LOGIC AND THEOREM PROVING workshop at TABLEAUX 2007
Abstract:
Abstract. We present a cut-free tableau calculus with histories and variables for
the EXPTIME-complete multi-modal logic of common knowledge (LCK). Our
calculus constructs the tableau using only one pass, so proof-search for testing
theoremhood of j does not exhibit the worst-case EXPTIME-behaviour for all j
as in two-pass methods. Our calculus also does not contain a “finitized w-ruleâ€
so that it detects cyclic branches as soon as they arise rather than by worst-case