AI RESEARCH
Catching a Moving Subspace: Low-Rank Bandits Beyond Stationarity
arXiv CS.LG
•
ArXi:2605.20269v1 Announce Type: new Many bandit deployments (recommendation, clinical dosing, ad targeting) share two facts prior work handles only in isolation: rewards live on a low-dimensional latent subspace, and that subspace drifts. Stationary low-rank bandits exploit rank but break under subspace change; non-stationary linear bandits adapt to drift but pay ambient rate $\widetilde{O}(d\sqrt{T