Hyperparameter importance and optimization of quantum neural networks across small datasets

被引:4
作者
Moussa, Charles [1 ]
Patel, Yash J. [1 ]
Dunjko, Vedran [1 ]
Baeck, Thomas [1 ]
van Rijn, Jan N. [1 ]
机构
[1] Leiden Univ, LIACS, Niels Bohrweg 1, NL-2333 CA Leiden, Netherlands
基金
荷兰研究理事会;
关键词
Hyperparameter importance; Quantum neural networks; Quantum machine learning; Hyperparameter optimization;
D O I
10.1007/s10994-023-06389-8
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
As restricted quantum computers become available, research focuses on finding meaningful applications. For example, in quantum machine learning, a special type of quantum circuit called a quantum neural network is one of the most investigated approaches. However, we know little about suitable circuit architectures or important model hyperparameters for a given task. In this work, we apply the functional ANOVA framework to the quantum neural network architectures to analyze which of the quantum machine learning hyperparameters are most influential for their predictive performance. We restrict our study to 7 open-source datasets from the OpenML-CC18 classification benchmark, which are small enough for simulations on quantum hardware with fewer than 20 qubits. Using this framework, three main levels of importance were identified, confirming expected patterns and revealing new insights. For instance, the learning rate is identified as the most important hyperparameter on all datasets, whereas the particular choice of entangling gates used is found to be the least important on all except for one dataset. In addition to identifying the relevant hyperparameters, for each of them, we also learned data-driven priors based on values that perform well on previously seen datasets, which can then be used to steer hyperparameter optimization processes. We utilize these priors in the hyperparameter optimization method hyperband and show that these improve performance against uniform sampling across all datasets by, on average, 0.53%, up to 6.11%, in cross-validation accuracy. We also demonstrate that such improvements hold on average regardless of the configuration hyperband is run with. Our work introduces new methodologies for studying quantum machine learning models toward quantum model selection in practice. All research code is made publicly available.
引用
收藏
页码:1941 / 1966
页数:26
相关论文
共 75 条
  • [1] [Anonymous], 1993, Math. Model. Comput. Exp.
  • [2] [Anonymous], 2015, PROC ICLR
  • [3] Parameterized quantum circuits as machine learning models
    Benedetti, Marcello
    Lloyd, Erika
    Sack, Stefan
    Fiorentini, Mattia
    [J]. QUANTUM SCIENCE AND TECHNOLOGY, 2019, 4 (04)
  • [4] Bergholm V., 2018, CoRR abs/1811.04968
  • [5] Biedenkapp A., 2018, INT C LEARNING INTEL, V11353, P115
  • [6] Biedenkapp A, 2017, AAAI CONF ARTIF INTE, P773
  • [7] CAVE: Configuration Assessment, Visualization and Evaluation
    Biedenkapp, Andre
    Marben, Joshua
    Lindauer, Marius
    Hutter, Frank
    [J]. LEARNING AND INTELLIGENT OPTIMIZATION, LION 12, 2019, 11353 : 115 - 130
  • [8] Bischl B, 2021, P NEURAL INFORM PROC
  • [9] Brazdil P., 2022, Metalearning: Applications to Automated Machine Learning and Data Mining, DOI [10.1007/978-3-030-67024-5, DOI 10.1007/978-3-030-67024-5]
  • [10] Random forests
    Breiman, L
    [J]. MACHINE LEARNING, 2001, 45 (01) : 5 - 32