English

Online Algorithm for Aggregating Experts' Predictions with Unbounded Quadratic Loss

Machine Learning 2025-01-14 v1

Abstract

We consider the problem of online aggregation of expert predictions with the quadratic loss function. We propose an algorithm for aggregating expert predictions which does not require a prior knowledge of the upper bound on the losses. The algorithm is based on the exponential reweighing of expert losses.

Cite

@article{arxiv.2501.06505,
  title  = {Online Algorithm for Aggregating Experts' Predictions with Unbounded Quadratic Loss},
  author = {Alexander Korotin and Vladimir V'yugin and Evgeny Burnaev},
  journal= {arXiv preprint arXiv:2501.06505},
  year   = {2025}
}
R2 v1 2026-06-28T21:03:25.209Z