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. 1010.3484
95
25

Hardness Results for Agnostically Learning Low-Degree Polynomial Threshold Functions

18 October 2010
Ilias Diakonikolas
Ryan O'Donnell
Rocco A. Servedio
Yi Wu
ArXiv (abs)PDFHTML
Abstract

Hardness results for maximum agreement problems have close connections to hardness results for proper learning in computational learning theory. In this paper we prove two hardness results for the problem of finding a low degree polynomial threshold function (PTF) which has the maximum possible agreement with a given set of labeled examples in Rn×{−1,1}.\R^n \times \{-1,1\}.Rn×{−1,1}. We prove that for any constants d≥1,\eps>0d\geq 1, \eps > 0d≥1,\eps>0, {itemize} Assuming the Unique Games Conjecture, no polynomial-time algorithm can find a degree-ddd PTF that is consistent with a (\half+\eps)(\half + \eps)(\half+\eps) fraction of a given set of labeled examples in Rn×{−1,1}\R^n \times \{-1,1\}Rn×{−1,1}, even if there exists a degree-ddd PTF that is consistent with a 1−\eps1-\eps1−\eps fraction of the examples. It is \NP\NP\NP-hard to find a degree-2 PTF that is consistent with a (\half+\eps)(\half + \eps)(\half+\eps) fraction of a given set of labeled examples in Rn×{−1,1}\R^n \times \{-1,1\}Rn×{−1,1}, even if there exists a halfspace (degree-1 PTF) that is consistent with a 1−\eps1 - \eps1−\eps fraction of the examples. {itemize} These results immediately imply the following hardness of learning results: (i) Assuming the Unique Games Conjecture, there is no better-than-trivial proper learning algorithm that agnostically learns degree-ddd PTFs under arbitrary distributions; (ii) There is no better-than-trivial learning algorithm that outputs degree-2 PTFs and agnostically learns halfspaces (i.e. degree-1 PTFs) under arbitrary distributions.

View on arXiv
Comments on this paper