A Short Note on a Variant of the Squint Algorithm

TL;DR

This paper introduces a simple variant of the Squint algorithm for the expert problem, providing a regret bound similar to recent algorithms, with a straightforward proof modification.

Contribution

It presents a new, simplified variant of the Squint algorithm and adapts its proof to establish comparable regret guarantees.

Findings

The variant achieves regret bounds similar to recent algorithms.

The proof modification is straightforward and elegant.

The approach simplifies analysis of the Squint algorithm.

Abstract

This short note describes a simple variant of the Squint algorithm of Koolen and Van Erven [2015] for the classic expert problem. Via an equally simple modification of their proof, we prove that this variant ensures a regret bound that resembles the one shown in a recent work by Freund et al. [2026] for a variant of the NormalHedge algorithm [Chaudhuri et al., 2009].