ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2002.08483
33
32

Strength from Weakness: Fast Learning Using Weak Supervision

19 February 2020
Joshua Robinson
Stefanie Jegelka
S. Sra
ArXivPDFHTML
Abstract

We study generalization properties of weakly supervised learning. That is, learning where only a few "strong" labels (the actual target of our prediction) are present but many more "weak" labels are available. In particular, we show that having access to weak labels can significantly accelerate the learning rate for the strong task to the fast rate of O(\nicefrac1n)\mathcal{O}(\nicefrac1n)O(\nicefrac1n), where nnn denotes the number of strongly labeled data points. This acceleration can happen even if by itself the strongly labeled data admits only the slower O(\nicefrac1n)\mathcal{O}(\nicefrac{1}{\sqrt{n}})O(\nicefrac1n​) rate. The actual acceleration depends continuously on the number of weak labels available, and on the relation between the two tasks. Our theoretical results are reflected empirically across a range of tasks and illustrate how weak labels speed up learning on the strong task.

View on arXiv
Comments on this paper