From Sequential Algorithm Selection to Parallel Portfolio Selection

被引:17
|
作者
Lindauer, M. [1 ]
Hoos, Holger H. [2 ]
Hutter, F. [1 ]
机构
[1] Univ Freiburg, D-79106 Freiburg, Germany
[2] Univ British Columbia, Vancouver, BC V5Z 1M9, Canada
来源
LEARNING AND INTELLIGENT OPTIMIZATION, LION 9 | 2015年 / 8994卷
关键词
Algorithm selection; Parallel portfolios; Constraint solving; Answer Set Programming; ANSWER; CONFIGURATION; SOLVER;
D O I
10.1007/978-3-319-19084-6_1
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In view of the increasing importance of hardware parallelism, a natural extension of per-instance algorithm selection is to select a set of algorithms to be run in parallel on a given problem instance, based on features of that instance. Here, we explore how existing algorithm selection techniques can be effectively parallelized. To this end, we leverage the machine learning models used by existing sequential algorithm selectors, such as 3S, ISAC, SATzilla and ME-ASP, and modify their selection procedures to produce a ranking of the given candidate algorithms; we then select the top n algorithms under this ranking to be run in parallel on n processing units. Furthermore, we adapt the pre-solving schedules obtained by aspeed to be effective in a parallel setting with different time budgets for each processing unit. Our empirical results demonstrate that, using 4 processing units, the best of our methods achieves a 12-fold average speedup over the best single solver on a broad set of challenging scenarios from the algorithm selection library.
引用
收藏
页码:1 / 16
页数:16
相关论文
共 50 条
  • [21] Online Parallel Portfolio Selection with Heterogeneous Island Model
    Balcar, Stepan
    Pilat, Martin
    2018 IEEE 30TH INTERNATIONAL CONFERENCE ON TOOLS WITH ARTIFICIAL INTELLIGENCE (ICTAI), 2018, : 757 - 764
  • [22] A Fast Parallel Selection Algorithm on GPUs
    Bakunas-Milanowski, Darius
    Rego, Vernon
    Sang, Janche
    Yu, Chansu
    2015 INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND COMPUTATIONAL INTELLIGENCE (CSCI), 2015, : 609 - 614
  • [23] Parallel genetic algorithm with fading selection
    Akopov, Andranik S.
    INTERNATIONAL JOURNAL OF COMPUTER APPLICATIONS IN TECHNOLOGY, 2014, 49 (3-4) : 325 - 331
  • [24] PORTFOLIO SELECTION
    Markowitz, Harry
    JOURNAL OF FINANCE, 1952, 7 (01): : 77 - 91
  • [25] Algorithm portfolio selection as a bandit problem with unbounded losses
    Gagliolo, Matteo
    Schmidhuber, Jurgen
    ANNALS OF MATHEMATICS AND ARTIFICIAL INTELLIGENCE, 2011, 61 (02) : 49 - 86
  • [26] Algorithm portfolio selection as a bandit problem with unbounded losses
    Matteo Gagliolo
    Jürgen Schmidhuber
    Annals of Mathematics and Artificial Intelligence, 2011, 61 : 49 - 86
  • [27] Portfolio selection with fuzzy synthetic evaluation and genetic algorithm
    Nayebpur, Hamid
    Bokaei, Mohsen Nazem
    ENGINEERING COMPUTATIONS, 2017, 34 (07) : 2422 - 2434
  • [28] Portfolio-Based Algorithm Selection for Circuit QBFs
    Hoos, Holger H.
    Peitl, Tomas
    Slivovsky, Friedrich
    Szeider, Stefan
    PRINCIPLES AND PRACTICE OF CONSTRAINT PROGRAMMING, 2018, 11008 : 195 - 209
  • [29] SIMPLE ALGORITHM FOR STONES VERSION OF PORTFOLIO SELECTION PROBLEM
    JUCKER, JV
    FARO, CD
    JOURNAL OF FINANCIAL AND QUANTITATIVE ANALYSIS, 1975, 10 (05) : 859 - 870
  • [30] A hybrid genetic quantitative algorithm for portfolio selection optimization
    Liu, ZD
    Proceedings of the 2005 International Conference on Management Science & Engineering (12th), Vols 1- 3, 2005, : 1834 - 1838