VOCEM：在 OffCEM 与 DR 之间求方差最优的离线策略评估估计量
Variance-Optimal Off-Policy Evaluation with Conjunct Effect Modeling
做离线策略评估的可以看看：VOCEM 用闭式系数在 OffCEM 和 DR 之间插值，23 个实验条件下方差都比两端低。
针对上下文老虎机策略的离线策略评估（OPE），动作级重要性加权会带来过高方差。此前提出的 OffCEM 用更稳定的簇级权重替代动作级权重，但代价是依赖奖励模型的局部正确性。论文证明在 DR 与 OffCEM 的共同假设下，存在一个连接两者的无偏估计量族，并提出 VOCEM，以闭式解给出总体最优插值系数，其方差不超过任一端点。在合成实验和两个大动作空间基准共 23 个条件中，VOCEM 全部优于 OffCEM 和 DR 两个端点。
Variance-Optimal Off-Policy Evaluation with Conjunct Effect Modeling
Off-policy evaluation (OPE) for contextual bandit policies becomes challenging when action-level importance weighting incurs excessive variance. Doubly robust (DR) estimation remains unbiased under common support but retains these high-variance action-level weights. A prior estimator, Off-policy evaluation with Conjunct Effect Model (OffCEM), replaces them with more stable cluster-level weights, at the cost of relying on local correctness of the reward model. In this paper, we show that, under the assumptions required by DR and OffCEM, there exists an unbiased family of estimators that interpolates between OffCEM and DR. Building on this result, we propose the Variance Optimal-CEM (VOCEM) estimator, which selects the interpolation coefficient to minimize variance. We derive the population-optimal coefficient in closed form and show that the resulting estimator has variance no larger than either endpoint, OffCEM or DR. Experiments in controlled synthetic settings and on two large-action benchmarks show that VOCEM improves upon both endpoints in all 23 evaluated conditions, exhibiting greater stability and empirical robustness.
