e8R0ytPhLv@OpenReview

Total: 1

#1 Eluder dimension: localise it! [PDF1] [Copy] [Kimi2] [REL]

Authors: Alireza Bakhtiari, Alex Ayoub, Samuel McLaughlin Robertson, David Janz, Csaba Szepesvari

We establish a lower bound on the eluder dimension in generalised linear model classes, showing that standard eluder dimension-based analysis cannot lead to first-order regret bounds. To address this, we introduce a localisation method for the eluder dimension; our analysis immediately recovers and improves on classic results for Bernoulli bandits, and allows for the first genuine first-order bounds for finite-horizon reinforcement learning tasks with bounded cumulative returns.

Subject: NeurIPS.2025 - Spotlight