Deep Clustering With a Constraint for Topological Invariance Based on Symmetric InfoNCE

Author:

Zhang Yuhui1,Wada Yuichiro23,Waida Hiroki4,Goto Kaito5,Hino Yusaku6,Kanamori Takafumi78

Affiliation:

1. Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552, Japan zhang.y.av@m.titech.ac.jp

2. Fujitsu Limited, Nakahara-ku, Kawasaki, Kanagawa 211-8588, Japan

3. RIKEN AIP, Chuo-ku, Tokyo 103-0027, Japan wada.yuichiro@jp.fujitsu.com

4. Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552, Japan waida.h.aa@m.titech.ac.jp

5. Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552, Japan kaitogoto7@gmail.com

6. Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552, Japan blueozk@gmail.com

7. Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552, Japan

8. RIKEN AIP, Chuo-ku, Tokyo 103-0027, Japan kanamori@c.titech.ac.jp

Abstract

Abstract We consider the scenario of deep clustering, in which the available prior knowledge is limited. In this scenario, few existing state-of-the-art deep clustering methods can perform well for both noncomplex topology and complex topology data sets. To address the problem, we propose a constraint utilizing symmetric InfoNCE, which helps an objective of the deep clustering method in the scenario of training the model so as to be efficient for not only noncomplex topology but also complex topology data sets. Additionally, we provide several theoretical explanations of the reason that the constraint can enhances the performance of deep clustering methods. To confirm the effectiveness of the proposed constraint, we introduce a deep clustering method named MIST, which is a combination of an existing deep clustering method and our constraint. Our numerical experiments via MIST demonstrate that the constraint is effective. In addition, MIST outperforms other state-of-the-art deep clustering methods for most of the commonly used 10 benchmark data sets.

Publisher

MIT Press

Subject

Cognitive Neuroscience,Arts and Humanities (miscellaneous)

Reference77 articles.

1. Geometry of q-exponential family of probability distributions;Amari;Entropy,2011

2. Self-labelling via simultaneous clustering and representation learning;Asano,2019

3. Non-parametric clustering using deep neural networks;Avgerinos;IEEE Access,2020

4. Connecting language and vision for natural language–based vehicle retrieval;Bai,2021

5. Rademacher and Gaussian complexities: Risk bounds and structural results;Bartlett;Journal of Machine Learning Research,2002

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