Automatic Feature Generation for Machine Learning Based Optimizing Compilation

被引:54
|
作者
Leather, Hugh [1 ]
Bonilla, Edwin [1 ]
O'Boyle, Michael [1 ]
机构
[1] Univ Edinburgh, Sch Informat, Edinburgh, Midlothian, Scotland
关键词
D O I
10.1109/CGO.2009.21
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Recent work has shown that machine learning can automate and in some cases outperform hand crafted compiler optimizations. Central to such an approach is that machine learning techniques typically rely upon summaries or features of the program. The quality of these features is critical to the accuracy of the resulting machine learned algorithm; no machine learning method will work well with poorly chosen features. However, due to the size and complexity of programs, theoretically there are an infinite number of potential features to choose from. The compiler writer now has to expend effort in choosing the best features from this space. This paper develops a novel mechanism to automatically find those features which most improve the quality of the machine learned heuristic. The feature space is described by a grammar and is then searched with genetic programming and predictive modeling. We apply this technique to loop unrolling in GCC 4.3.1 and evaluate our approach on a Pentium 6. On a benchmark suite of 57 programs, GCC's hard-coded heuristic achieves only 3% of the maximum performance available, while a state of the art machine learning approach with hand-coded features obtains 59%. Our feature generation technique is able to achieve 76% of the maximum available speedup, outperforming existing approaches.
引用
收藏
页码:81 / 91
页数:11
相关论文
共 50 条
  • [41] Automatic Layout Generation with Applications in Machine Learning Engine Evaluation
    Yang, Haoyu
    Chen, Wen
    Pathak, Piyush
    Gennari, Frank
    Lai, Ya-Chieh
    Yu, Bei
    [J]. 2019 ACM/IEEE 1ST WORKSHOP ON MACHINE LEARNING FOR CAD (MLCAD), 2019,
  • [42] Automatic Generation of a Portuguese Land Cover Map with Machine Learning
    Esteves, Antonio
    Valente, Nuno
    [J]. INTELLIGENT SYSTEMS AND APPLICATIONS, VOL 1, INTELLISYS 2023, 2024, 822 : 36 - 58
  • [43] On Automatic Generation of Training Images for Machine Learning in Automotive Applications
    Hsieh, Tong-Yu
    Lin, Yuan-Cheng
    Shen, Hsin-Yung
    [J]. 2019 IEEE INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE CIRCUITS AND SYSTEMS (AICAS 2019), 2019, : 225 - 228
  • [44] Automatic Generation of Program Affinity Policies Using Machine Learning
    Moore, Ryan W.
    Childers, Bruce R.
    [J]. COMPILER CONSTRUCTION, CC 2013, 2013, 7791 : 184 - 203
  • [45] Feature Selection Based on Extreme Learning Machine
    Wang, Zhaoxi
    Zhao, Meng
    Chen, Shengyong
    [J]. ICDLT 2019: 2019 3RD INTERNATIONAL CONFERENCE ON DEEP LEARNING TECHNOLOGIES, 2019, : 57 - 63
  • [46] Machine Learning for Feature-Based Analytics
    Wang, Li-C
    [J]. PROCEEDINGS OF THE 2018 INTERNATIONAL SYMPOSIUM ON PHYSICAL DESIGN (ISPD'18), 2018, : 74 - 81
  • [47] Optimizing intrusion detection using intelligent feature selection with machine learning model
    Aljehane, Nojood O.
    Mengash, Hanan A.
    Hassine, Siwar B. H.
    Alotaibi, Faiz A.
    Salama, Ahmed S.
    Abdelbagi, Sitelbanat
    [J]. ALEXANDRIA ENGINEERING JOURNAL, 2024, 91 : 39 - 49
  • [48] Optimizing Feature Selection for Solar Park Classification: Approaches with OBIA and Machine Learning
    Ladisa, Claudio
    Capolupo, Alessandra
    Tarantino, Eufemia
    [J]. COMPUTATIONAL SCIENCE AND ITS APPLICATIONS-ICCSA 2024 WORKSHOPS, PT V, 2024, 14819 : 286 - 301
  • [49] A Deep Learning Model Generation Method for Code Reuse and Automatic Machine Learning
    Lee, Keon Myung
    Hwang, Kyoung Soon
    Kim, Kwang Il
    Lee, Sang Hyun
    Park, Ki Sun
    [J]. PROCEEDINGS OF THE 2018 CONFERENCE ON RESEARCH IN ADAPTIVE AND CONVERGENT SYSTEMS (RACS 2018), 2018, : 47 - 52
  • [50] Automatic communication code generation in parallel compilation system
    Gong, Xue-rong
    Sheng, Yong-hong
    Zhang, Ping
    Lu, Lin-seng
    [J]. DCABES 2006 PROCEEDINGS, VOLS 1 AND 2, 2006, : 184 - 188