Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Short Note on a Variant of the Squint Algorithm

Haipeng Luo

TL;DR

This short note describes a simple variant of the Squint algorithm of Koolen and Van Erven [2015] and proves that this variant ensures a regret bound that resembles the one shown in a recent work by Freund et al.

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].

A Short Note on a Variant of the Squint Algorithm

TL;DR

This short note describes a simple variant of the Squint algorithm of Koolen and Van Erven [2015] and proves that this variant ensures a regret bound that resembles the one shown in a recent work by Freund et al.

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].
Paper Structure (5 sections, 4 theorems, 14 equations)

This paper contains 5 sections, 4 theorems, 14 equations.

Table of Contents

  1. Expert Problem and Squint
  2. Squint Algorithm and Guarantee
  3. A Variant of Squint
  4. Analysis
  5. Acknowledgment

Key Result

Lemma 1

For any $x \in [-1,1], R\in \mathbb{R}$ and $V \in \mathbb{R}^+$, we have

Theorems & Definitions (7)

  • Lemma 1
  • proof
  • Lemma 2: Lemma 1 of koolen2015second
  • proof
  • Lemma 3
  • proof
  • Theorem 4