We are experiencing minor difficulties with some of our submission forms, which we are working to resolve. You should still be able to submit items, but if you have any issues while submitting, please contact us at dspace-lib@mit.edu.

Deep Metric Learning via Facility Location

Thumbnail Image
Download
Name

1612.01213.pdf

Description
Accepted version
Size

5.53 MB

Format

Adobe PDF

Checksum (MD5)

4b41d0fb7f2584923d432d6da7657149

sword-2019-06-03T16:26:06.original.xml (130 B)
Original SWORD entry document
Author(s)
Song, Hyun Oh
•
Jegelka, Stefanie
•
Rathod, Vivek
•
Murphy, Kevin
Date Issued
July 2017
Publisher
IEEE
Citation
Song, Hyun Oh, Jegelka, Stefanie, Rathod, Vivek and Murphy, Kevin. 2017. "Deep Metric Learning via Facility Location."
Version
Author's final manuscript
Abstract
© 2017 IEEE. Learning image similarity metrics in an end-to-end fashion with deep networks has demonstrated excellent results on tasks such as clustering and retrieval. However, current methods, all focus on a very local view of the data. In this paper, we propose a new metric learning scheme, based on structured prediction, that is aware of the global structure of the embedding space, and which is designed to optimize a clustering quality metric (NMI). We show state of the art performance on standard datasets, such as CUB200-2011 [37], Cars196 [18], and Stanford online products [30] on NMI and R@K evaluation metrics.
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
10.1109/cvpr.2017.237
PrivacyPermissionsAccessibilityContact us
Notify us about copyright concerns.