The SMART approach to instance-optimal online learning

arXiv:2402.17720v1 Announce Type: new
Abstract: We devise an online learning algorithm -- titled Switching via Monotone Adapted Regret Traces (SMART) -- that adapts to the data and achieves regret that is instance optimal, i.e., simultaneously competitive on every input sequence compared to the performance of the follow-the-leader (FTL) policy and the worst case guarantee of any other input policy. We show that the regret of the SMART policy on any input sequence is within a multiplicative factor $e/(e-1) \approx 1.58$ …

abstract algorithm arxiv case cs.ds cs.it cs.lg data every instance leader math.it online learning performance policy smart traces type via