Popularity Bias Alignment Estimates

TL;DR

This paper extends the Popularity Bias Memorization theorem to more general degree distributions and provides bounds for alignment with top-k singular hyperspace, advancing theoretical understanding of popularity bias.

Contribution

It generalizes the theorem to arbitrary degree distributions and establishes both upper and lower bounds for hyperspace alignment.

Findings

Extended theorem to arbitrary degree distributions

Proved upper and lower bounds for hyperspace alignment

Enhanced theoretical framework for popularity bias analysis

Abstract

We are extending Popularity Bias Memorization theorem from arXiv:archive/2404.12008 in several directions. We extend it to arbitrary degree distributions and also prove both upper and lower estimates for the alignment with top-k singular hyperspace.