Arnaud Fietzke

LEO-II - A Cooperative Automatic Theorem Prover for Classical Higher-Order Logic (System Description)

LEO-II is a standalone, resolution-based higher-order theorem prover designed for effective cooperation with specialist provers for natural fragments of higher-order logic. At present LEO-II can cooperate with the first-order automated theorem provers E, SPASS, and Vampire. The improved performance of LEO-II, especially in comparison to its predecessor LEO, is due to several novel features including the exploitation of term sharing and term indexing techniques, support for primitive equality reasoning, and improved heuristics at the calculus level. LEO-II is implemented in Objective Caml and its problem representation language is the new TPTP THF language.

Superposition as a Decision Procedure for Timed Automata

The success of superposition-based theorem proving in first-order logic relies in particular on the fact that the superposition calculus can be turned into a decision procedure for various decidable fragments of first-order logic and has been successfully used to identify new decidable classes. In this paper, we extend this story to the hierarchic combination of linear arithmetic and first-order superposition. We show that decidability of reachability in timed automata can be obtained by instantiation of an abstract termination result for SUP(LA), the hierarchic combination of linear arithmetic and first-order superposition.

Labelled splitting

We define a superposition calculus with explicit splitting on the basis of labelled clauses. For the first time we show a superposition calculus with an explicit non-chronological backtracking rule sound and complete. The new backtracking rule advances backtracking with branch condensing known from SPASS. An experimental evaluation of an implementation of the new rule shows that it improves considerably on the previous SPASS splitting implementation. Finally, we discuss the relationship between labelled first-order splitting and DPLL style splitting with intelligent backtracking and clause learning.

Labelled Splitting

We define a superposition calculus with explicit splitting and an explicit, new backtracking rule on the basis of labelled clauses. For the first time we show a superposition calculus with explicit backtracking rule sound and complete. The new backtracking rule advances backtracking with branch condensing known from Spass . An experimental evaluation of an implementation of the new rule shows that it improves considerably the previous Spass splitting implementation. Finally, we discuss the relationship between labelled first-order splitting and DPLL style splitting with intelligent backtracking and clause learning.

Automatic generation of invariants for circular derivations in SUP(LA)

The hierarchic combination of linear arithmetic and first-order logic with free function symbols, FOL(LA), results in a strictly more expressive logic than its two parts. The SUP(LA) calculus can be turned into a decision procedure for interesting fragments of FOL(LA). For example, reachability problems for timed automata can be decided by SUP(LA) using an appropriate translation into FOL(LA). In this paper, we extend the SUP(LA) calculus with an additional inference rule, automatically generating inductive invariants from partial SUP(LA) derivations. The rule enables decidability of more expressive fragments, including reachability for timed automata with unbounded integer variables. We have implemented the rule in the SPASS(LA) theorem prover with promising results, showing that it can considerably speed up proof search and enable termination of saturation for practically relevant problems.