Handling incomplete heterogeneous data using VAEs

PATTERN RECOGNITION

卷:107

Article

Nazabal, Alfredo¹  Olmos, Pablo M.²  Ghahramani, Zoubin^3,4  Valera, Isabel^5,6 

[1] Alan Turing Inst, London, England

[2] Univ Carlos III, Madrid, Spain

[3] Univ Cambridge, Cambridge, England

[4] Uber AI Labs, San Francisco, CA USA

[5] Max Planck Inst Intelligent Syst, Tubingen, Germany

[6] Saarland Univ, Dept Comp Sci, Saarbrucken, Germany

关键词: Generative models;    Variational autoencoders;    Incomplete heterogenous data;   

DOI  :  10.1016/j.patcog.2020.107501

来源: Elsevier

【 摘 要 】

Variational autoencoders (VAEs), as well as other generative models, have been shown to be efficient and accurate for capturing the latent structure of vast amounts of complex high-dimensional data. However, existing VAEs can still not directly handle data that are heterogenous (mixed continuous and discrete) or incomplete (with missing data at random), which is indeed common in real-world applications. In this paper, we propose a general framework to design VAEs suitable for fitting incomplete heterogenous data. The proposed HI-VAE includes likelihood models for real-valued, positive real valued, interval, categorical, ordinal and count data, and allows accurate estimation (and potentially imputation) of missing data. Furthermore, HI-VAE presents competitive predictive performance in supervised tasks, outperforming supervised models when trained on incomplete data. (C) 2020 Elsevier Ltd. All rights reserved.

【 授权许可】

Free