Oracle-Efficient Hybrid Online Learning with Unknown Distribution
This provides the first oracle-efficient sublinear regret bounds for hybrid online learning with unknown distributions, confirming a prior conjecture and addressing computational efficiency in online learning for scenarios with mixed stochastic and adversarial elements.
The paper tackles the problem of oracle-efficient hybrid online learning with unknown feature distributions and adversarial labels, achieving sublinear regret bounds such as O~(T^{3/4}) for finite-VC classes and extending to shifting distributions and contextual bandits.
We study the problem of oracle-efficient hybrid online learning when the features are generated by an unknown i.i.d. process and the labels are generated adversarially. Assuming access to an (offline) ERM oracle, we show that there exists a computationally efficient online predictor that achieves a regret upper bounded by $\tilde{O}(T^{\frac{3}{4}})$ for a finite-VC class, and upper bounded by $\tilde{O}(T^{\frac{p+1}{p+2}})$ for a class with $α$ fat-shattering dimension $α^{-p}$. This provides the first known oracle-efficient sublinear regret bounds for hybrid online learning with an unknown feature generation process. In particular, it confirms a conjecture of Lazaric and Munos (JCSS 2012). We then extend our result to the scenario of shifting distributions with $K$ changes, yielding a regret of order $\tilde{O}(T^{\frac{4}{5}}K^{\frac{1}{5}})$. Finally, we establish a regret of $\tilde{O}((K^{\frac{2}{3}}(\log|\mathcal{H}|)^{\frac{1}{3}}+K)\cdot T^{\frac{4}{5}})$ for the contextual $K$-armed bandits with a finite policy set $\mathcal{H}$, i.i.d. generated contexts from an unknown distribution, and adversarially generated costs.