PreprintarXiv licencearXiv
On prediction from expert advice with more than five experts
Abstract
We prove that no single rank ordered adversary strategy is globally optimal for the prediction with expert advice problem with six or more experts, in both the geometric stopping and finite time horizon settings. The proof is based on establishing a leading order correction when one expert moves far ahead of the others. This allows us to connect optimal strategies between n and j<n experts and utilize recent results on the exact optimality set for the five expert problem.
Related
- Prediction with Expert Advice: Anytime Regret with Many Experts Matches the Fixed-Time Constant
- Prediction with Expert Advice under Heavy-Tailed Losses
- Abstention Spends Its Budget on Local Label Variance More Than on Novelty: Auditing Conformal Calibration and Selective Prediction under Molecular Scaffold Shift
- When Does Sparse Routing Help? The Geometry of Expert Separation in Mixture of Experts
- Fast Training of Mixture-of-Experts for Time Series Forecasting via Expert Loss Integration
BibTeX
@misc{calder2026prediction,
title = {{On prediction from expert advice with more than five experts}},
author = {Jeff Calder and Nadejda Drenska},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.05186},
eprint = {2610.05186},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.05186},
}
RIS
TY - UNPB
TI - On prediction from expert advice with more than five experts
AU - Jeff Calder
AU - Nadejda Drenska
PY - 2026
DA - 2026/10/04
PB - arXiv
DO - 10.48550/arxiv.2610.05186
UR - https://arxiv.org/abs/2610.05186
AB - We prove that no single rank ordered adversary strategy is globally optimal for the prediction with expert advice problem with six or more experts, in both the geometric stopping and finite time horizon settings. The proof is based on establishing a leading order correction when one expert moves far ahead of the others. This allows us to connect optimal strategies between n and j<n experts and utilize recent results on the exact optimality set for the five expert problem.
ER -