Exploration in Linear Bandits with Rich Action Sets and its Implications for Inference

TL;DR

This paper establishes a non-asymptotic lower bound on the eigenvalues of the design matrix in linear bandits with well-behaved action sets, revealing polynomial growth and implications for model selection and multi-agent clustering.

Contribution

It provides the first any-time lower bound on the design matrix eigenvalues for rich action spaces in linear bandits, extending previous asymptotic results to practical, finite-time scenarios.

Findings

Eigenvalues grow as (\u221a{n}) in well-behaved action spaces.

Epoch-based algorithms adapt exponentially to true model complexity.

No forced exploration needed for multi-agent clustering with spectral bounds.

Abstract

We present a non-asymptotic lower bound on the eigenspectrum of the design matrix generated by any linear bandit algorithm with sub-linear regret when the action set has well-behaved curvature. Specifically, we show that the minimum eigenvalue of the expected design matrix grows as $\Omega(\sqrt{n})$ whenever the expected cumulative regret of the algorithm is $O(\sqrt{n})$, where $n$ is the learning horizon, and the action-space has a constant Hessian around the optimal arm. This shows that such action-spaces force a polynomial lower bound rather than a logarithmic lower bound, as shown by \cite{lattimore2017end}, in discrete (i.e., well-separated) action spaces. Furthermore, while the previous result is shown to hold only in the asymptotic regime (as $n \to \infty$), our result for these "locally rich" action spaces is any-time. Additionally, under a mild technical assumption, we obtain a similar lower bound on the minimum eigen value holding with high probability. We apply our result to two practical scenarios -- \emph{model selection} and \emph{clustering} in linear bandits. For model selection, we show that an epoch-based linear bandit algorithm adapts to the true model complexity at a rate exponential in the number of epochs, by virtue of our novel spectral bound. For clustering, we consider a multi agent framework where we show, by leveraging the spectral result, that no forced exploration is necessary -- the agents can run a linear bandit algorithm and estimate their underlying parameters at once, and hence incur a low regret.