Generalization Performance of Some Learning Problems in Hilbert Functional Spaces

Free registration required

Executive Summary

This paper investigates the generalization performance of some learning problems in Hilbert functional Spaces. The paper introduces a notion of convergence of the estimated functional predictor to the best underlying predictor, and obtains an estimate on the rate of the convergence. This estimate allows one to derive generalization bounds on some learning formulations.

  • Format: PDF
  • Size: 93.8 KB