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. 2003.12895
18
15

Memorizing Gaussians with no over-parameterizaion via gradient decent on neural networks

28 March 2020
Amit Daniely
    VLM
    MLT
ArXivPDFHTML
Abstract

We prove that a single step of gradient decent over depth two network, with qqq hidden neurons, starting from orthogonal initialization, can memorize Ω(dqlog⁡4(d))\Omega\left(\frac{dq}{\log^4(d)}\right)Ω(log4(d)dq​) independent and randomly labeled Gaussians in Rd\mathbb{R}^dRd. The result is valid for a large class of activation functions, which includes the absolute value.

View on arXiv
Comments on this paper