MaSh: Machine Learning for Sledgehammer

Kühlwein, Daniel; Blanchette, Jasmin Christian; Kaliszyk, Cezary; Urban, Josef

doi:10.1007/978-3-642-39634-2_6

Cited by 54 publications

(58 citation statements)

References 28 publications

(38 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An earlier version of this work was presented at the ITP 2013 conference [23]. Since then, the naive Bayes algorithm has been ported from Python to Standard ML, to improve efficiency and reliability.…”

Section: Introductionmentioning

confidence: 99%

A Learning-Based Fact Selector for Isabelle/HOL

et al. 2016

Self Cite

View full text Add to dashboard Cite

Sledgehammer integrates automatic theorem provers in the proof assistant Isabelle/HOL. A key component, the fact selector, heuristically ranks the thousands of facts (lemmas, definitions, or axioms) available and selects a subset, based on syntactic similarity to the current proof goal. We introduce MaSh, an alternative that learns from successful proofs. New challenges arose from our "zero click" vision: MaSh integrates seamlessly with the users' workflow, so that they benefit from machine learning without having to install software, set up servers, or guide the learning. MaSh outperforms the old fact selector on large formalizations.

show abstract

Section: Introductionmentioning

confidence: 99%

A Learning-Based Fact Selector for Isabelle/HOL

et al. 2016

Self Cite

View full text Add to dashboard Cite

show abstract

“…This optimization may lose some constant features (e.g., if induced by a unification at a reduce step), however it very significantly speeds up the advising. Given that the features of the current path are f , the relevance of the eligible contrapositives (pre-selected by the lit indexing) is then computed according to the following modified naive-Bayes score (used by us for axiom selection in [11]):…”

Section: Learning-based Advising System and Related Infrastructurementioning

confidence: 99%

FEMaLeCoP: Fairly Efficient Machine Learning Connection Prover

Kaliszyk

Urban

2015

Logic for Programming, Artificial Intelligence, and Reasoning

Self Cite

View full text Add to dashboard Cite

Abstract. FEMaLeCoP is a connection tableau theorem prover based on leanCoP which uses efficient implementation of internal learningbased guidance for extension steps. Despite the fact that exhaustive use of such internal guidance can incur a significant slowdown of the raw inferencing process, FEMaLeCoP trained on related proofs can prove many problems that cannot be solved by leanCoP. In particular on the MPTP2078 benchmark, FEMaLeCoP adds 90 (15.7%) more problems to the 574 problems that are provable by leanCoP. FEMaLeCoP is thus the first AI/ATP system convincingly demonstrating that guiding the internal inference algorithms of theorem provers by knowledge learned from previous proofs can significantly improve the performance of the provers. This paper describes the system, discusses the technology developed, and evaluates the system.

show abstract

“…As common in such evaluations [6,8,12] the most complementary methods are computed by a greedy algorithm, and the resulting greedy sequences are shown from top to bottom in the table. The total improvement is in this case 6.2%, i.e.…”

mentioning

confidence: 99%

Lemmatization for Stronger Reasoning in Large Theories

Kaliszyk¹,

Urban²,

VyskoăźIl³

2015

Frontiers of Combining Systems

Self Cite

View full text Add to dashboard Cite

Abstract. In this work we improve ATP performance in large theories by the reuse of lemmas derived in previous related problems. Given a large set of related problems to solve, we run automated theorem provers on them, extract a large number of lemmas from the proofs found and post-process the lemmas to make them usable in the remaining problems. Then we filter the lemmas by several tools and extract their proof dependencies, and use machine learning on such proof dependencies to add the most promising generated lemmas to the remaining problems.On such enriched problems we run the automated provers again, solving more problems. We describe this method and the techniques we used, and measure the improvement obtained. On the MPTP2078 large-theory benchmark the method yields 6.6% and 6.2% more problems proved in two different evaluation modes.

show abstract

MaSh: Machine Learning for Sledgehammer

Cited by 54 publications

References 28 publications

A Learning-Based Fact Selector for Isabelle/HOL

A Learning-Based Fact Selector for Isabelle/HOL

FEMaLeCoP: Fairly Efficient Machine Learning Connection Prover

Lemmatization for Stronger Reasoning in Large Theories

Contact Info

Product

Resources

About