Nearly Optimal Algorithms for Piecewise-Stationary Cascading Bandits

TL;DR

This paper introduces nearly optimal algorithms for piecewise-stationary cascading bandits, effectively detecting change points and adapting to evolving user preferences with improved regret bounds and minimal tuning.

Contribution

The paper proposes two new algorithms using a generalized likelihood ratio test for change detection in non-stationary cascading bandits, achieving near-optimal regret bounds and fewer tuning parameters.

Findings

Regret bounds of order O(√NLT log T) for the proposed algorithms.

Algorithms are nearly optimal, matching the lower bound up to a logarithmic factor.

Numerical experiments confirm the effectiveness of the algorithms on real and synthetic data.

Abstract

Cascading bandit (CB) is a popular model for web search and online advertising, where an agent aims to learn the $K$ most attractive items out of a ground set of size $L$ during the interaction with a user. However, the stationary CB model may be too simple to apply to real-world problems, where user preferences may change over time. Considering piecewise-stationary environments, two efficient algorithms, \texttt{GLRT-CascadeUCB} and \texttt{GLRT-CascadeKL-UCB}, are developed and shown to ensure regret upper bounds on the order of $\mathcal{O}(\sqrt{NLT\log{T}})$, where $N$ is the number of piecewise-stationary segments, and $T$ is the number of time slots. At the crux of the proposed algorithms is an almost parameter-free change-point detector, the generalized likelihood ratio test (GLRT). Comparing with existing works, the GLRT-based algorithms: i) are free of change-point-dependent information for choosing parameters; ii) have fewer tuning parameters; iii) improve at least the $L$ dependence in regret upper bounds. In addition, we show that the proposed algorithms are optimal (up to a logarithm factor) in terms of regret by deriving a minimax lower bound on the order of $\Omega(\sqrt{NLT})$ for piecewise-stationary CB. The efficiency of the proposed algorithms relative to state-of-the-art approaches is validated through numerical experiments on both synthetic and real-world datasets.