天津科技大学学报2025，Vol.40Issue(2)：61-70,10.DOI:10.13364/j.issn.1672-6510.20240023
基于监督对比正则化项的信息蒸馏生成对抗网络
Information Distillation Generative Adversarial Net Based on Supervised Contrastive Regularization
摘要
Abstract
For generative adversarial net(GAN),traditional methods mainly maximize disentangling the latent presentation based on the mutual information between the disentangled representation and the generated data,but rarely analyze the inde-pendence among dimensions of the latent vector.In this article,we propose an information distillation generative adversarial net based on supervised contrastive regularization(IDGAN-SC).The IDGAN-SC model firstly learns disentangled represen-tation space through training β-VAE,which enforces strong correlation between the disentangled representation space and the generative model.Then,the model constructs the disentangled structure by maximizing the mutual information between the disentangled latent vectors and the generated data.Furthermore,the model utilizes the contrastive classification information of the supervised contrastive regularization to enhance the independence between dimensions of the latent vectors.In our present study,we conducted quantitative and qualitative experiments on the dSprites,MNIST,and CelebA datasets.Experi-ments showed that IDGAN-SC significantly outperformed current disentanglement methods based on the disentanglement metrics.关键词
生成对抗网络/变分自编码器/解耦表示/对比学习Key words
generative adversarial net/variational auto-encoder/disentangled representation/contrastive learning分类
计算机与自动化引用本文复制引用
陈亚瑞,王晓捷,李晴,刘浩天,史艳翠,赵婷婷..基于监督对比正则化项的信息蒸馏生成对抗网络[J].天津科技大学学报,2025,40(2):61-70,10.基金项目
国家自然科学基金项目(61976156) （61976156）
