Jan Jakubruv, Karel Chvalovsk'y, Z. Goertzel, C. Kaliszyk, Mirek Olvs'ak, Bartosz Piotrowski, S. Schulz, M. Suda, J. Urban
{"title":"MizAR 60 for Mizar 50","authors":"Jan Jakubruv, Karel Chvalovsk'y, Z. Goertzel, C. Kaliszyk, Mirek Olvs'ak, Bartosz Piotrowski, S. Schulz, M. Suda, J. Urban","doi":"10.48550/arXiv.2303.06686","DOIUrl":null,"url":null,"abstract":"As a present to Mizar on its 50th anniversary, we develop an AI/TP system that automatically proves about 60\\% of the Mizar theorems in the hammer setting. We also automatically prove 75\\% of the Mizar theorems when the automated provers are helped by using only the premises used in the human-written Mizar proofs. We describe the methods and large-scale experiments leading to these results. This includes in particular the E and Vampire provers, their ENIGMA and Deepire learning modifications, a number of learning-based premise selection methods, and the incremental loop that interleaves growing a corpus of millions of ATP proofs with training increasingly strong AI/TP systems on them. We also present a selection of Mizar problems that were proved automatically.","PeriodicalId":296683,"journal":{"name":"International Conference on Interactive Theorem Proving","volume":"47 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Conference on Interactive Theorem Proving","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2303.06686","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
As a present to Mizar on its 50th anniversary, we develop an AI/TP system that automatically proves about 60\% of the Mizar theorems in the hammer setting. We also automatically prove 75\% of the Mizar theorems when the automated provers are helped by using only the premises used in the human-written Mizar proofs. We describe the methods and large-scale experiments leading to these results. This includes in particular the E and Vampire provers, their ENIGMA and Deepire learning modifications, a number of learning-based premise selection methods, and the incremental loop that interleaves growing a corpus of millions of ATP proofs with training increasingly strong AI/TP systems on them. We also present a selection of Mizar problems that were proved automatically.