ID 原文 译文
5234 根据整合后的新 2K 序列生成满足差分隐私的社交网络发布图。 Finally, a social network graph satisfy-ing differential privacy was generated according to the new 2K sequences after integration.
5235 基于真实数据集，通过模拟仿真将所提方案与其他经典方案进行比较， Based on the real datasets, thescheme was compared with the classical schemes through simulation experiments.
5236 实验结果表明，所提方案在保证较强隐私保护性的同时，提高了发布数据的准确性和可用性。 The results demonstrate that it im-proves the accuracy and usability of data while ensuring desired privacy protection level.
5237 在连续动作任务中，深度强化学习通常采用高斯分布作为策略函数。 In continuous action tasks, deep reinforcement learning usually uses Gaussian distribution as a policy function.
5238 针对高斯分布策略函数由于截断动作导致算法收敛速度变慢的问题，提出了一种重要性采样优势估计器（ISAE）。 Aiming at the problem that the Gaussian distribution policy function slows down due to the clipped action, an importancesampling advantage estimator was proposed.
5239 该估计器在通用优势估计器（GAE）的基础上，引入了重要性采样机制，通过计算边界动作的目标策略与行动策略比率修正截断动作带来的值函数偏差，提高了算法的收敛速度。 Based on the general advantage estimator, an importance sampling mechan-ism was introduced by the estimator to improve the convergence speed of the algorithm and correct the deviation of thevalue function caused by calculating the target strategy and action strategy ratio of the boundary action.
5240 此外，ISAE 引入了 L 参数，通过限制重要性采样率的范围，提高了样本的可靠度，保证了网络参数的稳定。 In addition, the Lparameter was introduced by ISAE which improved the reliability of the sample and limited the stability of the networkparameters by limiting the range of the importance sampling rate.
5241 为了验证 ISAE 的有效性，将 ISAE 与近端策略优化结合并与其他算法在 MuJoCo 平台上进行比较。实验结果表明，ISAE 具有更快的收敛速度。 In order to verify the effectiveness of the ISAE, apply-ing it to proximal policy optimization and comparing it with other algorithms on the MuJoCo platform. Experimental re-sults show that ISAE has a faster convergence rate.
5242 无连接接入协议允许机器类终端不需要建立无线承载就进行小数据传输，以降低系统信令开销与终端能耗。 Connectionless access allows massive machine type communication (mMTC) devices to transmit small pack-ets without establishment of radio bearers, significantly reducing device power consumption and control signaling over-head.
5243 两阶段无连接接入协议可通过合理分配调度请求阶段和数据传输阶段的资源提高系统吞吐量，特别适合中高负载场景。 Two-stage connectionless access (TSCLA) improves throughput and resource efficiency by optimally allocatingresources between scheduling request (SR) phase and data transmission phase, which can be used for bigger packets andhigh traffic load.