Interpretability for Neural Networks from the Perspective of Probability Density

被引:0
|
作者
Lu, Lu [1 ]
Pan, Tingting [1 ]
Zhao, Junhong [1 ]
Yang, Jie [1 ]
机构
[1] Dalian Univ Technol, Sch Math Sci, Dalian, Peoples R China
基金
中国国家自然科学基金;
关键词
neural networks; interpretability; probability density; gaussian distribution;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Currently, most of works about interpretation of neural networks are to visually explain the features learned by hidden layers. This paper explores the relationship between the input units and the output units of neural network from the perspective of probability density. For classification problems, it shows that the probability density function (PDF) of the output unit can be expressed as a mixture of three Gaussian density functions whose mean and variance are related to the information of the input units, under the assumption that the input units are independent of each other and obey a Gaussian distribution. The experimental results show that the theoretical distribution of the output unit is basically consistent with the actual distribution.
引用
收藏
页码:1502 / 1507
页数:6
相关论文
共 50 条
  • [1] Interpretability of Neural Networks with Probability Density Functions
    Pan, Tingting
    Pedrycz, Witold
    Cui, Jiahui
    Yang, Jie
    Wu, Wei
    [J]. ADVANCED THEORY AND SIMULATIONS, 2022, 5 (03)
  • [2] New Perspective of Interpretability of Deep Neural Networks
    Kimura, Masanari
    Tanaka, Masayuki
    [J]. 2020 3RD INTERNATIONAL CONFERENCE ON INFORMATION AND COMPUTER TECHNOLOGIES (ICICT 2020), 2020, : 78 - 85
  • [3] On the interpretability of quantum neural networks
    Pira, Lirande
    Ferrie, Chris
    [J]. QUANTUM MACHINE INTELLIGENCE, 2024, 6 (02)
  • [4] On the determination of probability density functions by using Neural Networks
    Garrido, L
    Juste, A
    [J]. COMPUTER PHYSICS COMMUNICATIONS, 1998, 115 (01) : 25 - 31
  • [5] Probability density estimation using artificial neural networks
    Likas, A
    [J]. COMPUTER PHYSICS COMMUNICATIONS, 2001, 135 (02) : 167 - 175
  • [6] A probability density function generator based on neural networks
    Chen, Chi-Hua
    Song, Fangying
    Hwang, Feng-Jang
    Wu, Ling
    [J]. PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS, 2020, 541
  • [7] Estimate of a Probability Density Function through Neural Networks
    Reyneri, Leonardo
    Colla, Valentina
    Vannucci, Marco
    [J]. ADVANCES IN COMPUTATIONAL INTELLIGENCE, IWANN 2011, PT I, 2011, 6691 : 57 - 64
  • [8] On Interpretability of Artificial Neural Networks: A Survey
    Fan, Feng-Lei
    Xiong, Jinjun
    Li, Mengzhou
    Wang, Ge
    [J]. IEEE TRANSACTIONS ON RADIATION AND PLASMA MEDICAL SCIENCES, 2021, 5 (06) : 741 - 760
  • [9] Explainable Neural Networks: Achieving Interpretability in Neural Models
    Chakraborty, Manomita
    [J]. ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING, 2024, 31 (06) : 3535 - 3550
  • [10] Conditional probability density function estimation with sigmoidal neural networks
    Sarajedini, A
    Hecht-Nielsen, R
    Chau, PM
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1999, 10 (02): : 231 - 238