考虑延迟的智能选择算法，提升冷冻电镜数据采集效率
Latency-Aware Contextual Bandit: Application to Cryo-EM Data Collection
- 引入延迟感知上下文博弈框架，动态调整可选操作集
- 在电影推荐和冷冻电镜数据上，累积收益显著优于基线
- 适合需权衡响应时间与决策质量的实时系统应用
我们提出一种延迟感知的上下文博弈框架，该框架扩展了标准上下文博弈问题，允许学习者在动作延迟条件下自适应选择动作并切换决策集。学习者观察上下文后可从决策集中选择多个动作，总耗时由所选子集决定。该问题可建模为半马尔可夫决策过程（SMDP）的特例，其中上下文和延迟来自未知分布。基于贝尔曼最优方程，我们设计了上下文在线动作过滤（COAF）算法，平衡探索、利用与动作延迟，以最小化相对于最优平均回报策略的遗憾。我们对算法进行分析，证明其遗憾上界与上下文博弈文献中的经典结果一致。在电影推荐数据集和冷冻电镜（cryo-EM）数据上的数值实验表明，该方法能高效最大化时间维度上的累积收益。
原文摘要 · Abstract (English)
We introduce a latency-aware contextual bandit framework that generalizes the standard contextual bandit problem, where the learner adaptively selects arms and switches decision sets under action delays. In this setting, the learner observes the context and may select multiple arms from a decision set, with the total time determined by the selected subset. The problem can be framed as a special case of semi-Markov decision processes (SMDPs), where contexts and latencies are drawn from an unknown distribution. Leveraging the Bellman optimality equation, we design the contextual online arm filtering (COAF) algorithm, which balances exploration, exploitation, and action latency to minimize regret relative to the optimal average-reward policy. We analyze the algorithm and show that its regret upper bounds match established results in the contextual bandit literature. In numerical experiments on a movie recommendation dataset and cryogenic electron microscopy (cryo-EM) data, we demonstrate that our approach efficiently maximizes cumulative reward over time.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。