Heavy-Tailed Linear Bandits: Huber Regression with One-Pass Update

We study the stochastic linear bandits with heavy-tailed noise. Two principled strategies for handling heavy-tailed noise, truncation and median-of-means, have been introduced to heavy-tailed bandits. Nonetheless, these methods rely on specific noise assumptions or bandit structures, limiting their applicability to general settings. The recent work [Huang et al.2024] develops a soft truncation method via the adaptive Huber regression to address these limitations. However, their method suffers undesired computational costs: it requires storing all historical data and performing a full pass over these data at each round. In this paper, we propose a one-pass algorithm based on the online mirror descent framework. Our method updates using only current data at each round, reducing the per-round computational cost from O(t log T) to O(1) with respect to current round t and the time horizon T, and achieves a near-optimal and variance-aware regret of order O(d T1-ε/2(1+ε) √∑ₜ₌₁T νₜ² + d T1-ε/2(1+ε)) where d is the dimension and νₜ1+ε is the (1+ε)-th central moment of reward at round t.