Parameter-Free Dynamic Regret for Unconstrained Linear Bandits

Alberto Rumi, Andrew Jacobsen, Nicolò Cesa-Bianchi, Fabio Vitale

TL;DR: We achieve the first optimal dynamic regret guarantee in unconstrained linear bandits, without requiring prior knowledge of comparator variability.

Abstract: We study dynamic regret minimization in unconstrained adversarial linear bandit problems. In this setting, a learner must minimize the cumulative loss relative to an arbitrary sequence of comparators $\boldsymbol{u}_1,\ldots,\boldsymbol{u}_T$ in $\mathbb{R}^d$, but receives only *point-evaluation feedback* on each round. We provide a simple approach to combining the guarantees of several bandit algorithms, allowing us to optimally adapt to the number of switches $S_T = \sum_t\mathbb{I}\\{\boldsymbol{u}\_t \neq \boldsymbol{u}\_{t-1}\\}$ of an arbitrary comparator sequence. In particular, we provide the *first* algorithm for linear bandits achieving the optimal regret guarantee of order $\mathcal{O}\big(\sqrt{d(1+S_T) T}\big)$ up to poly-logarithmic terms *without prior knowledge of $S_T$*, thus resolving a long-standing open problem.

Code Dataset Promise: No

Format Confirmation: I agree that I have read and followed the formatting instructions for the camera ready version.

Submission Number: 1343