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. 2106.15000
13
18

Optimal Convergence Rates for the Orthogonal Greedy Algorithm

28 June 2021
Jonathan W. Siegel
Jinchao Xu
ArXivPDFHTML
Abstract

We analyze the orthogonal greedy algorithm when applied to dictionaries D\mathbb{D}D whose convex hull has small entropy. We show that if the metric entropy of the convex hull of D\mathbb{D}D decays at a rate of O(n−12−α)O(n^{-\frac{1}{2}-\alpha})O(n−21​−α) for α>0\alpha > 0α>0, then the orthogonal greedy algorithm converges at the same rate on the variation space of D\mathbb{D}D. This improves upon the well-known O(n−12)O(n^{-\frac{1}{2}})O(n−21​) convergence rate of the orthogonal greedy algorithm in many cases, most notably for dictionaries corresponding to shallow neural networks. These results hold under no additional assumptions on the dictionary beyond the decay rate of the entropy of its convex hull. In addition, they are robust to noise in the target function and can be extended to convergence rates on the interpolation spaces of the variation norm. We show empirically that the predicted rates are obtained for the dictionary corresponding to shallow neural networks with Heaviside activation function in two dimensions. Finally, we show that these improved rates are sharp and prove a negative result showing that the iterates generated by the orthogonal greedy algorithm cannot in general be bounded in the variation norm of D\mathbb{D}D.

View on arXiv
Comments on this paper