In this paper, we consider a remote inference system, where a neural network is used to infer a time-varying target (e.g., robot movement), based on features (e.g., video clips) that are progressively received from a sensing node (e.g., a camera). Each feature is a temporal sequence of sensory data. The learning performance of the system is determined by (i) the timeliness and (ii) the temporal sequence length of the features, where we use Age of Information (AoI) as a metric for timeliness. While a longer feature can typically provide better learning performance, it often requires more channel resources for sending the feature. To minimize the time-averaged inference error, we study a learning and communication co-design problem that jointly optimizes feature length selection and transmission scheduling. When there is a single sensor-predictor pair and a single channel, we develop low-complexity optimal co-designs for both the cases of time-invariant and time-variant feature length. When there are multiple sensor-predictor pairs and multiple channels, the co-design problem becomes a restless multi-arm multi-action bandit problem that is PSPACE-hard. For this setting, we design a low-complexity algorithm to solve the problem. Trace-driven evaluations suggest that the proposed co-designs can significantly reduce the time-averaged inference error of remote inference systems.

翻译：本文研究了一种远程推理系统，其中利用神经网络基于从感知节点（如摄像头）逐步接收的特征（如视频片段）来推断时变目标（如机器人运动）。每个特征是一段时间序列的感知数据。系统的学习性能由（i）特征的及时性和（ii）时间序列长度决定，其中我们使用信息年龄（AoI）作为及时性的度量指标。较长的特征通常能提供更好的学习性能，但往往需要更多的信道资源来传输该特征。为最小化时间平均推理误差，我们研究了一个学习与通信协同设计问题，该问题联合优化特征长度选择和传输调度。针对单一传感器-预测器对和单一信道的场景，我们分别针对时不变和时变特征长度这两种情况，设计了低复杂度的最优协同方案。当存在多个传感器-预测器对和多个信道时，该协同设计问题转化为一个难以处理的易逝多臂多动作摇臂问题（PSPACE-hard）。针对这一设定，我们设计了一种低复杂度算法来求解该问题。基于轨迹驱动的评估表明，所提出的协同设计方案能够显著降低远程推理系统的时间平均推理误差。