arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2511.13060cs.LGcs.AI
收藏

在线决策中的延迟与顺序效应

Latency and Ordering Effects in Online Decisions

Duo Yi

AI总结:

针对在线决策中存在的延迟反馈与顺序敏感非交换动态问题,本文基于Bregman散度推导超额损失结构化下界,扩展至非凸场景并给出项估计监测方案,为真实系统调优提供可解释框架。

摘要:

在线决策系统通常在延迟反馈和顺序敏感(非交换)动态下运行:行动会影响观测结果的到达情况与到达顺序。我们以Bregman散度$D_Φ$作为损失基准,证明超额基准损失存在结构化下界$L \ge L_{\mathrm{ideal}} + g_1(λ) + g_2(\varepsilon_\star) + g_{12}(λ,\varepsilon_\star) - D_{\mathrm{ncx}}$,其中$g_1$和$g_2$是针对延迟和顺序敏感性的校准惩罚项,$g_{12}$刻画二者的几何交互作用,$D_{\mathrm{ncx}}\ge0$是在凸勒让德假设下会消失的非凸性/近似惩罚项。我们将该不等式扩展到prox-正则和弱凸场景,获得了超出凸场景的鲁棒性保证。我们还给出了通过简单的2×2随机实验和流式诊断指标(有效样本量、裁剪率、交互热力图)估计与监测这四项的操作方案。该框架将异质延迟、非交换性和实现差距效应整合为单一可解释的下界表述,可在真实系统中进行压力测试与调优。

英文摘要:

Online decision systems routinely operate under delayed feedback and order-sensitive (noncommutative) dynamics: actions affect which observations arrive, and in what sequence. Taking a Bregman divergence $D_Φ$ as the loss benchmark, we prove that the excess benchmark loss admits a structured lower bound $L \ge L_{\mathrm{ideal}} + g_1(λ) + g_2(\varepsilon_\star) + g_{12}(λ,\varepsilon_\star) - D_{\mathrm{ncx}}$, where $g_1$ and $g_2$ are calibrated penalties for latency and order-sensitivity, $g_{12}$ captures their geometric interaction, and $D_{\mathrm{ncx}}\ge 0$ is a nonconvexity/approximation penalty that vanishes under convex Legendre assumptions. We extend this inequality to prox-regular and weakly convex settings, obtaining robust guarantees beyond the convex case. We also give an operational recipe for estimating and monitoring the four terms via simple $2\times 2$ randomized experiments and streaming diagnostics (effective sample size, clipping rate, interaction heatmaps). The framework packages heterogeneous latency, noncommutativity, and implementation-gap effects into a single interpretable lower-bound statement that can be stress-tested and tuned in real-world systems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑