Multi-Level Variational Autoencoder: Learning Disentangled Representations From Grouped Observations

Author:

Bouchacourt Diane,Tomioka Ryota,Nowozin Sebastian

Abstract

We would like to learn a representation of the data that reflects the semantics behind a specific grouping of the data, where within a group the samples share a common factor of variation. For example, consider a set of face images grouped by identity. We wish to anchor the semantics of the grouping into a disentangled representation that we can exploit. However, existing deep probabilistic models often assume that the samples are independent and identically distributed, thereby disregard the grouping information. We present the Multi-Level Variational Autoencoder (ML-VAE), a new deep probabilistic model for learning a disentangled representation of grouped data. The ML-VAE separates the latent representation into semantically relevant parts by working both at the group level and the observation level, while retaining efficient test-time inference. We experimentally show that our model (i) learns a semantically meaningful disentanglement, (ii) enables control over the latent representation, and (iii) generalises to unseen groups.

Publisher

Association for the Advancement of Artificial Intelligence (AAAI)

Subject

General Medicine

Cited by 39 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Contrastive Disentanglement for Quantitative Ultrasound Muscle Atrophy Evaluation;IEEE Transactions on Biomedical Engineering;2024-08

2. Soft-Attention Based Person Re-Identification in Real-world Settings using Variational AutoEncoders;2024 16th International Conference on Human System Interaction (HSI);2024-07-08

3. DSDRNet: Disentangling Representation and Reconstruct Network for Domain Generalization;2024 International Joint Conference on Neural Networks (IJCNN);2024-06-30

4. A Survey on Variational Autoencoders in Recommender Systems;ACM Computing Surveys;2024-06-24

5. PoseRec: 3D Human Pose Driven Online Advertisement Recommendation for Micro-videos;Proceedings of the 2024 International Conference on Multimedia Retrieval;2024-05-30

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3