A Classification Supervised Auto-Encoder Based on Predefined Evenly-Distributed Class Centroids
This is an incremental improvement for machine learning tasks requiring enhanced autoencoder models with integrated classification.
The paper tackles the problem of improving autoencoder performance by introducing a classification supervised autoencoder (CSAE) based on predefined evenly-distributed class centroids (PEDCC), which achieved comprehensive optimal results in encoding, decoding, classification, and generalization without specifying concrete numbers.
Classic variational autoencoders are used to learn complex data distributions, that are built on standard function approximators. Especially, VAE has shown promise on a lot of complex task. In this paper, a new autoencoder model - classification supervised autoencoder (CSAE) based on predefined evenly-distributed class centroids (PEDCC) is proposed. Our method uses PEDCC of latent variables to train the network to ensure the maximization of inter-class distance and the minimization of inner-class distance. Instead of learning mean/variance of latent variables distribution and taking reparameterization of VAE, latent variables of CSAE are directly used to classify and as input of decoder. In addition, a new loss function is proposed to combine the loss function of classification. Based on the basic structure of the universal autoencoder, we realized the comprehensive optimal results of encoding, decoding, classification, and good model generalization performance at the same time. Theoretical advantages are reflected in experimental results.