Error Exponent in Agnostic PAC Learning

arXiv:2405.00792v1 Announce Type: new
Abstract: Statistical learning theory and the Probably Approximately Correct (PAC) criterion are the common approach to mathematical learning theory. PAC is widely used to analyze learning problems and algorithms, and have been studied thoroughly. Uniform worst case bounds on the convergence rate have been well established using, e.g., VC theory or Radamacher complexity. However, in a typical scenario the performance could be much better. In this paper, we consider PAC learning using a somewhat different tradeoff, …

abstract algorithms analyze arxiv case convergence criterion cs.lg error rate statistical stat.ml theory type uniform