We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CV

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computer Vision and Pattern Recognition

Title: Self-Supervised Learning by Estimating Twin Class Distributions

Abstract: We present TWIST, a novel self-supervised representation learning method by classifying large-scale unlabeled datasets in an end-to-end way. We employ a siamese network terminated by a softmax operation to produce twin class distributions of two augmented images. Without supervision, we enforce the class distributions of different augmentations to be consistent. In the meantime, we regularize the class distributions to make them sharp and diverse. Specifically, we minimize the entropy of the distribution for each sample to make the class prediction for each sample assertive and maximize the entropy of the mean distribution to make the predictions of different samples diverse. In this way, TWIST can naturally avoid the trivial solutions without specific designs such as asymmetric network, stop-gradient operation, or momentum encoder. Different from the clustering-based methods which alternate between clustering and learning, our method is a single learning process guided by a unified loss function. As a result, TWIST outperforms state-of-the-art methods on a wide range of tasks, including unsupervised classification, linear classification, semi-supervised learning, transfer learning, and some dense prediction tasks such as detection and segmentation. Codes and pre-trained models are given on: this https URL
Comments: Technical report
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Cite as: arXiv:2110.07402 [cs.CV]
  (or arXiv:2110.07402v3 [cs.CV] for this version)

Submission history

From: Feng Wang [view email]
[v1] Thu, 14 Oct 2021 14:39:39 GMT (14971kb,D)
[v2] Fri, 15 Oct 2021 03:09:41 GMT (14971kb,D)
[v3] Fri, 22 Oct 2021 09:55:54 GMT (14971kb,D)

Link back to: arXiv, form interface, contact.