Discriminative Auto-Encoder With Local and Global Graph Embedding

Abstract

In order to exploit the potential intrinsic low-dimensional structure of the high-dimensional data from the manifold learning perspective, we propose a global graph embedding with globality-preserving property, which requires that samples should be mapped close to their low-dimensional class representation data distribution centers in the embedding space. Then we propose a novel local and global graph embedding auto-encoder(LGAE) to capture the geometric structure of data, its cost function have three terms, a reconstruction loss to reproduce the input data based on the learned representation, a local graph embedding regularization to enforce mapping the neighboring samples close together in the embedding space, a global embedding regularization to enforce mapping samples close to their low-dimensional class representation distribution centers. Thus in the learning process, our LGAE can map samples from same class close together in the embedding space, as well as reduce the scatter within-class and increase the margin between-class, it will also detect the local and global intrinsic geometric structure of data and discover the latent discriminant information in the embedding space. We build stacked LGAE for classification tasks and conduct comprehensive experiments on several benchmark datasets, the results confirm that our proposed framework can learn discriminative representation, speed up the network convergence process, and significantly improve the classification performance. INDEX TERMS Manifold learning, graph embedding, discriminative auto-encoder, deep learning.

High‐Resolution Range Profile Sequence Recognition Based on Transformer with Temporal–Spatial Fusion and Label Smoothing

“…Different classifier confidence levels were defined according to the distance between classifiers given by the confusion matrix; then the values and posterior probabilities of the support vector machines were integrated into the basic probability assignments to achieve a recognition method that combines support vector machines with the evidence theory. Li et al [ 32 ] proposed an extreme learning machine (ELM)‐based recognition method by introducing the L21 norm to reduce the undesirable effects of data noise points and outliers, which made the ELM model more stable. The machine learning method achieves automatic feature extraction, while the above methods extract features by considering HRRP as a whole, which ignores the correlation and temporal information between HRRPs, resulting in information loss.…”

Fault Diagnosis of Electric Motors Using Deep Learning Algorithms and Its Application: A Review

“…Unlike shallow neural networks, deep learning models can directly use the original data as input and learn data features layer-by-layer through a multilayer model, thus resulting in more effective feature extraction [17]. Currently, deep belief networks (DBN) [18][19][20][21][22][23][24][25][26][27][28][29][30][31][32], autoencoders (AE) [33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52], convolutional neural networks (CNN) [53][54][55][56][57][58][59][60], and recurrent neural networks (RNN) [61][62][63][64][65]…”

“…A common three-layer unsupervised feature learning model is the autoencoder (AE). The output can be restored to the input as closely as feasible using adaptive learning features [33][34][35]. The corresponding autoencoding network model has evolved according to different standards for defining feature expression, such as sparsity features, noise reduction features, regular constraint features, and so on.…”

“…where L(x, y) is the reconstruction error; λΩ is the weight attenuation term to prevent overfitting [35]; m, n l , and S l represent the number of samples, the number of network layers, and the number of neurons in the l layer, respectively; W ij represents the weight of the interlayer connection; b is the unit bias of the l layer.…”