计算机应用 ›› 2020, Vol. 40 ›› Issue (9): 2665-2669.DOI: 10.11772/j.issn.1001-9081.2020010011

• 先进计算 • 上一篇    下一篇

基于启发式搜索的浮点表达式设计空间探索方法

李钊, 董霄霄, 黄程程, 任崇广   

  1. 山东理工大学 计算机科学与技术学院, 山东 淄博 255000
  • 收稿日期:2020-01-14 修回日期:2020-03-02 出版日期:2020-09-10 发布日期:2020-03-11
  • 通讯作者: 李钊
  • 作者简介:李钊(1983-),男,山东淄博人,讲师,博士,主要研究方向:近似计算、机器学习;董霄霄(1996-),女,山东济宁人,硕士研究生,主要研究方向:深度学习;黄程程(1997-),男,四川仁寿人,硕士研究生,主要研究方向:深度学习;任崇广(1982-),男,山东临沂人,副教授,博士,主要研究方向:大数据、云计算。
  • 基金资助:
    国家自然科学基金资助项目(61701286);山东省自然科学基金资助项目(ZR2018LF002,ZR2017LF004);山东省高等学校青年创新团队发展计划(2019KJN048);淄博市校城融合项目(2018ZBXC021)。

Design space exploration method for floating-point expression based on heuristic search

LI Zhao, DONG Xiaoxiao, HUANG Chengcheng, REN Chongguang   

  1. School of Computer Science and Technology, Shandong University of Technology, Zibo Shandong 255000, China
  • Received:2020-01-14 Revised:2020-03-02 Online:2020-09-10 Published:2020-03-11
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61701286), the Natural Science Foundation of Shandong Province (ZR2018LF002, ZR2017LF004), the Development Plan of Youth Innovation Teams in Higher Educations of Shandong Province (2019KJN048), the University and City Integration Project of Zibo City (2018ZBXC021).

摘要: 为了提高浮点表达式设计空间的探索效率,提出一种基于启发搜索的浮点表达式设计空间探索方法。在每次迭代过程中首先对非支配表达式的设计空间进行探索,同时将非支配表达式和可支配表达式分别添加到非支配列表和可支配列表中。当迭代完成后对可支配列表中的表达式进行探索,从中选择非支配的表达式,并对其邻域进行探索。将新的非支配表达式添加到非支配列表中,有效提高了非支配表达式的多样性和随机性。最后再次对非支配列表进行探索,得到最终的等价表达式,并进一步提高最优表达式的性能。与现有的浮点表达式设计空间的探索方法相比较,所提出的方法使计算精度提高了2%~9%,并减少了5%~19%的计算时间和4%~7%的资源消耗。实验结果表明,该方法可有效提高空间探索效率。

关键词: 浮点表达式, 设计空间探索, 启发式搜索, 计算精度, 计算时间, 资源消耗

Abstract: In order to improve the exploration efficiency of the design space for floating-point expression, a design space exploration method based on heuristic search was proposed. The design space of non-dominated expression was explored firstly during each iteration. At the same time, the non-dominated expression and the dominated expression were added to the non-dominated list and the dominated list respectively. Then the expression in the dominated list was explored after the iteration, the non-dominated expression in the dominated list was selected, and the neighborhood of the non-dominated expression in the dominated list was explored. And the new non-dominated expression was added to the non-dominated list, effectively improving the diversity and randomness of the non-dominated expression. Finally, the non-dominated list was explored again to obtain the final equivalent expression and further improve the performance of optimal expression. Compared with the existing design space exploration methods for floating-point expression, the proposed method has the calculation accuracy increased by 2% to 9%, the calculation time reduced by 5% to 19% and the resource consumption reduced by 4% to 7%. Experimental results show that the proposed method can effectively improve the efficiency of design space exploration.

Key words: floating-point expression, design space exploration, heuristic search, calculation accuracy, calculation time, resource consumption

中图分类号: