Asymptotic Supervised Predictive Classifiers under Partition Exchangeability

TL;DR

This paper proves that under partition exchangeability, supervised predictive classifiers converge asymptotically with large data, allowing simpler classifiers to replace more complex ones without loss of accuracy.

Contribution

It establishes the asymptotic equivalence of simultaneous and marginal classifiers under partition exchangeability in supervised learning.

Findings

Predictive classifiers converge with infinite data

Differences between classifiers become negligible

Simpler classifiers can replace complex ones in large data regimes

Abstract

The convergence of simultaneous and marginal predictive classifiers under partition exchangeability in supervised classification is obtained. The result shows the asymptotic convergence of these classifiers under infinite amount of training or test data, such that after observing umpteen amount of data, the differences between these classifiers would be negligible. This is an important result from the practical perspective as under the presence of sufficiently large amount of data, one can replace the simpler marginal classifier with computationally more expensive simultaneous one.