MisConv: Convolutional Neural Networks for Missing Data

被引:2
|
作者
Likowski, Marcin Przewiez [1 ]
Smieja, Marek [1 ]
Struski, Lukasz [1 ]
Tabor, Jacek [1 ]
机构
[1] Jagiellonian Univ, Fac Math & Comp Sci, 6 Lojasiewicza St, PL-30348 Krakow, Poland
关键词
CHAINED EQUATIONS; IMPUTATION; MACHINE; VALUES;
D O I
10.1109/WACV51458.2022.00297
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Processing of missing data by modern neural networks, such as CNNs, remains a fundamental, yet unsolved challenge, which naturally arises in many practical applications, like image inpainting or autonomous vehicles and robots. While imputation-based techniques are still one of the most popular solutions, they frequently introduce unreliable information to the data and do not take into account the uncertainty of estimation, which may be destructive for a machine learning model. In this paper, we present MisConv, a general mechanism, for adapting various CNN architectures to process incomplete images. By modeling the distribution of missing values by the Mixture of Factor Analyzers, we cover the spectrum of possible replacements and find an analytical formula for the expected value of convolution operator applied to the incomplete image. The whole framework is realized by matrix operations, which makes MisConv extremely efficient in practice. Experiments performed on various image processing tasks demonstrate that MisConv achieves superior or comparable performance to the state-of-the-art methods.
引用
收藏
页码:2917 / 2926
页数:10
相关论文
共 50 条
  • [41] Impact of missing data in evaluating artificial neural networks trained on complete data
    Markey, MK
    Tourassi, GD
    Margolis, M
    DeLong, DM
    [J]. COMPUTERS IN BIOLOGY AND MEDICINE, 2006, 36 (05) : 516 - 525
  • [42] Towards Effective Classification of Imbalanced Data with Convolutional Neural Networks
    Raj, Vidwath
    Magg, Sven
    Wermter, Stefan
    [J]. ARTIFICIAL NEURAL NETWORKS IN PATTERN RECOGNITION, 2016, 9896 : 150 - 162
  • [43] Circular Convolutional Neural Networks for Panoramic Images and Laser Data
    Schubert, Stefan
    Neubert, Peer
    Poschmann, Johannes
    Pretzel, Peter
    [J]. 2019 30TH IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV19), 2019, : 653 - 660
  • [44] Dynamic Sampling in Convolutional Neural Networks for Imbalanced Data Classification
    Pouyanfar, Samira
    Tao, Yudong
    Mohan, Anup
    Tian, Haiman
    Kaseb, Ahmed S.
    Gauen, Kent
    Dailey, Ryan
    Aghajanzadeh, Sarah
    Lu, Yung-Hsiang
    Chen, Shu-Ching
    Shyu, Mei-Ling
    [J]. IEEE 1ST CONFERENCE ON MULTIMEDIA INFORMATION PROCESSING AND RETRIEVAL (MIPR 2018), 2018, : 112 - 117
  • [45] Traffic Data Imputation Using Deep Convolutional Neural Networks
    Benkraouda, Ouafa
    Thodi, Bilal Thonnam
    Yeo, Hwasoo
    Menendez, Monica
    Jabari, Saif Eddin
    [J]. IEEE ACCESS, 2020, 8 : 104740 - 104752
  • [46] Weather Data For The Prevention Of Agricultural Production With Convolutional Neural Networks
    Tarik, Hajji
    Jamil, Ouazzani Mohemmad
    [J]. 2019 INTERNATIONAL CONFERENCE ON WIRELESS TECHNOLOGIES, EMBEDDED AND INTELLIGENT SYSTEMS (WITS), 2019,
  • [47] Convolutional neural networks for signal detection in real LIGO data
    Zelenka, Ondrej
    Bruegmann, Bernd
    Ohme, Frank
    [J]. PHYSICAL REVIEW D, 2024, 110 (02)
  • [48] Convolutional neural networks applied to data organized as OLAP cubes
    Caputo, Rodrigo Ribeiro
    dos Santos, Edimilson Batista
    da Rocha, Leonardo Chaves Dutra
    [J]. EXPERT SYSTEMS, 2023, 40 (10)
  • [49] The role of data embedding in equivariant quantum convolutional neural networks
    Das, Sreetama
    Martina, Stefano
    Caruso, Filippo
    [J]. Quantum Machine Intelligence, 2024, 6 (02)
  • [50] Precise Approximation of Convolutional Neural Networks for Homomorphically Encrypted Data
    Lee, Junghyun
    Lee, Eunsang
    Lee, Joon-Woo
    Kim, Yongjune
    Kim, Young-Sik
    No, Jong-Seon
    [J]. IEEE ACCESS, 2023, 11 : 62062 - 62076