Hammering Mizar by Learning Clause Guidance

We describe a very large improvement of existing hammer-style proof automation over large ITP libraries by combining learning and theorem proving. In particular, we have integrated state-of-the-art machine learners into the E automated theorem prover, and developed methods that allow learning and efficient internal guidance of E over the whole Mizar library. The resulting trained system improves the real-time performance of E on the Mizar library by 70% in a single-strategy setting.

[1]  Thibault Gauthier,et al.  TacticToe: Learning to Reason with HOL4 Tactics , 2017, LPAR.

[2]  Cezary Kaliszyk,et al.  Reinforcement Learning of Theorem Proving , 2018, NeurIPS.

[3]  Cezary Kaliszyk,et al.  Random Forests for Premise Selection , 2015, FroCos.

[4]  Cezary Kaliszyk,et al.  FEMaLeCoP: Fairly Efficient Machine Learning Connection Prover , 2015, LPAR.

[5]  Josef Urban,et al.  Enhancing ENIGMA Given Clause Guidance , 2018, CICM.

[6]  Chih-Jen Lin,et al.  LIBLINEAR: A Library for Large Linear Classification , 2008, J. Mach. Learn. Res..

[7]  Cezary Kaliszyk,et al.  Efficient Semantic Features for Automated Reasoning over Large Theories , 2015, IJCAI.

[8]  Rajeev Raman,et al.  SEPIA: Search for Proofs Using Inferred Automata , 2015, CADE.

[9]  Josef Urban,et al.  ENIGMA-NG: Efficient Neural and Gradient-Boosted Inference Guidance for E , 2019, CADE.

[10]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[11]  Josef Urban,et al.  ProofWatch: Watchlist Guidance for Large Theories in E , 2018, ITP.

[12]  Cezary Kaliszyk,et al.  A Learning-Based Fact Selector for Isabelle/HOL , 2016, Journal of Automated Reasoning.

[13]  Josef Urban,et al.  ProofWatch Meets ENIGMA: First Experiments , 2018 .

[14]  Stephan Schulz,et al.  System Description: E 1.8 , 2013, LPAR.

[15]  Josef Urban,et al.  BliStrTune: hierarchical invention of theorem proving strategies , 2017, CPP.

[16]  Josef Urban,et al.  MPTP 0.2: Design, Implementation, and Initial Experiments , 2006, Journal of Automated Reasoning.

[17]  Adam Naumowicz,et al.  Mizar in a Nutshell , 2010, J. Formaliz. Reason..

[18]  Josef Urban,et al.  DeepMath - Deep Sequence Models for Premise Selection , 2016, NIPS.

[19]  Cezary Kaliszyk,et al.  Deep Network Guided Proof Search , 2017, LPAR.

[20]  LPAR-21, 21st International Conference on Logic for Programming, Artificial Intelligence and Reasoning, Maun, Botswana, May 7-12, 2017 , 2017, LPAR.

[21]  Stephan Schulz,et al.  E - a brainiac theorem prover , 2002, AI Commun..

[22]  Cezary Kaliszyk,et al.  MizAR 40 for Mizar 40 , 2013, Journal of Automated Reasoning.

[23]  Josef Urban,et al.  MaLARea SG1- Machine Learner for Automated Reasoning with Semantic Guidance , 2008, IJCAR.

[24]  Jesse Alama,et al.  Premise Selection for Mathematics by Corpus Analysis and Kernel Methods , 2011, Journal of Automated Reasoning.

[25]  Cezary Kaliszyk,et al.  Monte Carlo Tableau Proof Search , 2017, CADE.

[26]  R. Polikar,et al.  Ensemble based systems in decision making , 2006, IEEE Circuits and Systems Magazine.

[27]  Andrei Voronkov,et al.  First-Order Theorem Proving and Vampire , 2013, CAV.

[28]  Josef Urban,et al.  ENIGMA: Efficient Learning-Based Inference Guiding Machine , 2017, CICM.

[29]  Cezary Kaliszyk,et al.  Learning-Assisted Automated Reasoning with Flyspeck , 2012, Journal of Automated Reasoning.

[30]  Cezary Kaliszyk,et al.  Hammering towards QED , 2016, J. Formaliz. Reason..

[31]  Josef Urban,et al.  MaLeCoP Machine Learning Connection Prover , 2011, TABLEAUX.