Dimensionality's blessing: Clustering images by underlying distribution
Many high dimensional vector distances tend to a constant. This is typically considered a negative “contrastloss” phenomenon that hinders clustering and other machine learning techniques. We reinterpret “contrast-loss” as a blessing. Re-deriving “contrast-loss” using the law of large numbers, we sho...
Saved in:
Main Authors: | LIN, Wen-yan, LAI, Jian-Huang, LIU, Siying, MATSUSHITA, Yasuyuki |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2018
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/4810 https://ink.library.smu.edu.sg/context/sis_research/article/5813/viewcontent/Lin_Dimensionalitys_Blessing_Clustering_CVPR_2018_paper.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
Similar Items
-
Aligning images in the wild
by: LIN, Wen-yan, et al.
Published: (2012) -
EditAnything: Empowering unparalleled flexibility in image editing and generation
by: GAO, Shanghua, et al.
Published: (2023) -
LPT: Long-tailed prompt tuning for image classification
by: DONG, Bowen, et al.
Published: (2023) -
Semi-supervised hierarchical clustering for personalized web image organization
by: MENG, Lei, et al.
Published: (2012) -
Exploring category-agnostic clusters for open-set domain adaptation
by: PAN, Yingwei, et al.
Published: (2020)