Autoencoder in Autoencoder Networks

• Published in: IEEE transaction on neural networks and learning systems Vol. 35; no. 2; pp. 2263 - 2275
• Main Authors: Zhang, Changqing, Geng, Yu, Han, Zongbo, Liu, Yeqing, Fu, Huazhu, Hu, Qinghua
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.02.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Bidirectional encoding; Coding; Complementarity; complete representation; Correlation; Data mining; Degradation; Encoding; Feature extraction; Information processing; Kernel; Machine learning; multiview representation learning; Networks; Representation learning; Representations
• ISSN: 2162-237X; 2162-2388; 2162-2388
• DOI: 10.1109/TNNLS.2022.3189239

Modeling complex correlations on multiview data is still challenging, especially for high-dimensional features with possible noise. To address this issue, we propose a novel unsupervised multiview representation learning (UMRL) algorithm, termed autoencoder in autoencoder networks (AE2-Nets). The proposed framework effectively encodes information from high-dimensional heterogeneous data into a compact and informative representation with the proposed bidirectional encoding strategy. Specifically, the proposed AE2-Nets conduct encoding in two directions: the inner-AE-networks extract view-specific intrinsic information (forward encoding), while the outer-AE-networks integrate this view-specific intrinsic information from different views into a latent representation (backward encoding). For the nested architecture, we further provide a probabilistic explanation and extension from hierarchical variational autoencoder. The forward-backward strategy flexibly addresses high-dimensional (noisy) features within each view and encodes complementarity across multiple views in a unified framework. Extensive results on benchmark datasets validate the advantages compared to the state-of-the-art algorithms.