Corpus ID: 221534503

Hyperparameter Optimization via Sequential Uniform Designs

  title={Hyperparameter Optimization via Sequential Uniform Designs},
  author={Zebin Yang and Aijun Zhang},
  • Zebin Yang, Aijun Zhang
  • Published 2020
  • Computer Science, Mathematics
  • ArXiv
  • Hyperparameter tuning or optimization plays a central role in the automated machine learning (AutoML) pipeline. It is a challenging task as the response surfaces of hyperparameters are generally unknown, and the evaluation of each experiment is expensive. In this paper, we reformulate hyperparameter optimization as a kind of computer experiment and propose a novel sequential uniform design (SeqUD) for hyperparameter optimization. It is advantageous as a) it adaptively explores the… CONTINUE READING


    Publications referenced by this paper.
    Hyperopt: a Python library for model selection and hyperparameter optimization
    • 205
    Practical Bayesian Optimization of Machine Learning Algorithms
    • 3,277
    • Highly Influential
    • PDF
    Multi-Task Bayesian Optimization
    • 356
    • PDF
    Collaborative hyperparameter tuning
    • 185
    • PDF
    Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimization
    • 469
    • PDF
    Efficient Global Optimization of Expensive Black-Box Functions
    • 4,668
    • PDF
    Tunability: Importance of Hyperparameters of Machine Learning Algorithms
    • 73
    • PDF
    Model selection for support vector machines via uniform design
    • 144
    • PDF
    Sequential Model-Based Optimization for General Algorithm Configuration
    • 1,379
    • PDF