Speeding up Common Hyperparameter Optimization Methods by a Two-Phase-Search

被引:0
|
作者
Wendt, Alexander [1 ]
Wuschnig, Marco [1 ]
Lechner, Martin [1 ]
机构
[1] TU Vienna, ICT, Christian Doppler Lab Embedded Machine Learning, Vienna, Austria
关键词
hyperparameter; machine learning; support vector machine; random forest; Bayesian optimization; optimization;
D O I
10.1109/iecon43393.2020.9254801
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Hyperparameter search concerns everybody who works with machine learning. We compare publicly available hyperparameter searches on four datasets. We develop metrics to measure the performance of hyperparameter searches across datasets of different sizes as well as machine learning algorithms. Further, we propose a method of speeding up the search by using subsets of data. Results show that random search performs well compared to Bayesian methods and that a combined search can speed up the search by a factor of 5.
引用
收藏
页码:517 / 522
页数:6
相关论文
共 44 条
  • [1] Speeding up the Hyperparameter Optimization of Deep Convolutional Neural Networks
    Hinz, Tobias
    Navarro-Guerrero, Nicolas
    Magg, Sven
    Wermter, Stefan
    INTERNATIONAL JOURNAL OF COMPUTATIONAL INTELLIGENCE AND APPLICATIONS, 2018, 17 (02)
  • [2] Speeding up dynamic search methods in speech recognition
    Gosztolya, G
    Kocsor, A
    INNOVATIONS IN APPLIED ARTIFICIAL INTELLIGENCE, 2005, 3533 : 98 - 100
  • [3] Speeding Up Automatic Hyperparameter Optimization of Deep Neural Networks by Extrapolation of Learning Curves
    Domhan, Tobias
    Springenberg, Jost Tobias
    Hutter, Frank
    PROCEEDINGS OF THE TWENTY-FOURTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI), 2015, : 3460 - 3468
  • [4] TransBO: Hyperparameter Optimization via Two-Phase Transfer Learning
    Li, Yang
    Shen, Yu
    Jiang, Huaijun
    Zhang, Wentao
    Yang, Zhi
    Zhang, Ce
    Cui, Bin
    PROCEEDINGS OF THE 28TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, KDD 2022, 2022, : 956 - 966
  • [5] Speeding Up Chemical Searches Using the Inverted Index: The Convergence of Chemoinformatics and Text Search Methods
    Nasr, Ramzi
    Vernica, Rares
    Li, Chen
    Baldi, Pierre
    JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2012, 52 (04) : 891 - 900
  • [6] Data structures for speeding up Tabu Search when solving sparse quadratic unconstrained binary optimization problems
    Liang, Ricardo N.
    Anacleto, Eduardo A. J.
    Meneses, Claudio N.
    JOURNAL OF HEURISTICS, 2022, 28 (04) : 433 - 479
  • [7] Data structures for speeding up Tabu Search when solving sparse quadratic unconstrained binary optimization problems
    Ricardo N. Liang
    Eduardo A. J. Anacleto
    Cláudio N. Meneses
    Journal of Heuristics, 2022, 28 : 433 - 479
  • [8] Speeding Up Multi-Objective Hyperparameter Optimization by Task Similarity-Based Meta-Learning for the Tree-Structured Parzen Estimator
    Watanabe, Shuhei
    Awad, Noor
    Onishi, Masaki
    Hutter, Frank
    PROCEEDINGS OF THE THIRTY-SECOND INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2023, 2023, : 4380 - 4388
  • [9] Speeding up the two-phase PT-flash, with applications for calculation of miscible displacement
    Michael, ML
    FLUID PHASE EQUILIBRIA, 1998, 143 (1-2) : 1 - 12
  • [10] Two-step spectral library pre-search: A novel approach for speeding up compound identification
    Zhu, Qiang
    Yu, Jiancheng
    Hu, Junjack
    Ding, Chuanfan
    INTERNATIONAL JOURNAL OF MASS SPECTROMETRY, 2017, 417 : 40 - 44