JAIST Repository >
b. 情報科学研究科・情報科学系 >
b30. リサーチレポート >
Research Report - School of Information Science : ISSN 0918-7553 >
IS-RR-2013 >

このアイテムの引用には次の識別子を使用してください: http://hdl.handle.net/10119/11349

タイトル: raSAT: SMT for Polynomial Inequality
著者: To, Van Khanh
Ogawa, Mizuhito
キーワード: interval arithmetic
affine arithmetic
polynomial constraints
発行日: 2013-05-27
出版者: 北陸先端科学技術大学院大学情報科学研究科
誌名: Research report (School of Information Science, Japan Advanced Institute of Science and Technology)
巻: IS-RR-2013-003
開始ページ: 1
終了ページ: 23
抄録: This paper presents an iterative approximation refinement, called raSAT loop, which solves polynomial inequality constraints on real numbers. The approximation scheme consists of interval arithmetic (over-approximation, aiming to decide unsatisfiability) and testing (underapproximation, aiming to decide satisfiability). If both of them fail to decide, input intervals are refined by decomposition. raSAT loop is implemented as an SMT raSAT with miniSAT 2.2 as a backend SAT solver. Experiments including simple benchmarks for estimating effects of input measures (i.e., degrees, number of variables, and number of constraints) and QF_NRA benchmarks from SMT-LIB show that raSAT is comparable to Z3 4.3, and sometimes outperforms, especially with high degree of polynomials.
URI: http://hdl.handle.net/10119/11349
資料タイプ: publisher


ファイル 記述 サイズ形式
IS-RR-2013-003.pdf598KbAdobe PDF見る/開く



お問い合わせ先 : 北陸先端科学技術大学院大学 研究推進課図書館情報係