A Classification Supervised Auto-Encoder Based on Predefined Evenly-Distributed Class Centroids

TL;DR

This paper introduces a classification supervised autoencoder (CSAE) that uses predefined evenly-distributed class centroids to enhance class separation and improve encoding, decoding, and classification simultaneously.

Contribution

The paper proposes a novel CSAE model utilizing PEDCC to directly classify latent variables without reparameterization, improving class separation and overall autoencoder performance.

Findings

Enhanced inter-class distance and intra-class compactness.

Simultaneous optimization of encoding, decoding, and classification.

Theoretical advantages confirmed by experimental results.

Abstract

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.