VC dimension bounds for product unit networks

被引:2
|
作者
Schmitt, M [1 ]
机构
[1] Ruhr Univ Bochum, Fak Math, Lehrstuhl Math & Informat, D-44780 Bochum, Germany
来源
IJCNN 2000: PROCEEDINGS OF THE IEEE-INNS-ENNS INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOL IV | 2000年
关键词
D O I
10.1109/IJCNN.2000.860767
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
A product unit is a formal neuron that multiplies its input values instead of summing them. Furthermore, it has weights acting as exponents instead of being factors. We investigate the complexity of learning for networks containing product units. We establish bounds on the Vapnik-Chervonenkis (VC) dimension that can be used to assess the generalization capabilities of these networks. In particular, we show that the VC dimension for these networks is not larger than the best known bound for sigmoidal networks. For higher-order networks we derive upper bounds that are independent of the degree of these networks. We also contrast these results with lower bounds.
引用
收藏
页码:165 / 170
页数:6
相关论文
共 50 条
  • [21] Tight Lower Bounds on the VC-dimension of Geometric Set Systems
    Csikos, Monika
    Mustafa, Nabil H.
    Kupayskii, Audrey
    JOURNAL OF MACHINE LEARNING RESEARCH, 2019, 20
  • [22] Nearly Optimal VC-Dimension and Pseudo-Dimension Bounds for Deep Neural Network Derivatives
    Yang, Yahong
    Yang, Haizhao
    Xiang, Yang
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [23] Tight bounds on the maximum size of a set of permutations with bounded VC-dimension
    Cibulka, Josef
    Kyncl, Jan
    JOURNAL OF COMBINATORIAL THEORY SERIES A, 2012, 119 (07) : 1461 - 1478
  • [24] Dimension independent bounds for general shallow networks
    Mhaskar, H. N.
    NEURAL NETWORKS, 2020, 123 : 142 - 152
  • [25] Neural networks with local receptive fields and superlinear VC dimension
    Schmitt, M
    NEURAL COMPUTATION, 2002, 14 (04) : 919 - 956
  • [26] VC dimension of Graph Neural Networks with Pfaffian activation functions
    D'Inverno, Giuseppe Alessio
    Bianchini, Monica
    Scarselli, Franco
    NEURAL NETWORKS, 2025, 182
  • [27] On using the Poincare polynomial for calculating the VC dimension of neural networks
    Schmitt, M
    NEURAL NETWORKS, 2001, 14 (10) : 1465 - 1465
  • [28] BOUNDS ON THE SAMPLE COMPLEXITY OF BAYESIAN LEARNING USING INFORMATION-THEORY AND THE VC DIMENSION
    HAUSSLER, D
    KEARNS, M
    SCHAPIRE, RE
    MACHINE LEARNING, 1994, 14 (01) : 83 - 113
  • [29] On the Upper Bounds of Fractional Metric Dimension of Symmetric Networks
    Javaid, Muhammad
    Aslam, Muhammad Kamran
    Liu, Jia-Bao
    JOURNAL OF MATHEMATICS, 2021, 2021
  • [30] Radial basis function neural networks have superlinear VC dimension
    Schmitt, M
    COMPUTATIONAL LEARNING THEORY, PROCEEDINGS, 2001, 2111 : 14 - 30