As data traffic volume continues to increase, caching of popular content at strategic network locations closer to the end user can enhance not only user experience but ease the utilization of highly congested links in the network. A key challenge in the area of proactive caching is finding the optimal locations to host the popular content items under various optimization criteria. These problems are combinatorial in nature and therefore finding optimal and/or near optimal decisions is computationally expensive. In this paper a framework is proposed to reduce the computational complexity of the underlying integer mathematical program by first predicting decision variables related to optimal locations using a deep convolutional neural network (CNN). The CNN is trained in an offline manner with optimal solutions and is then used to feed a much smaller optimization problems which is amenable for real-time decision making. Numerical investigations reveal that the proposed approach can provide in an online manner high quality decision making; a feature which is crucially important for real-world implementations.

中文翻译：

---

通过模型驱动和基于人工智能技术的协同作用在移动网络中进行网络编排

随着数据流量的不断增加，在靠近最终用户的战略网络位置缓存流行内容不仅可以增强用户体验，还可以简化网络中高度拥塞的链接的利用率。主动缓存领域的一个关键挑战是在各种优化标准下找到托管流行内容项目的最佳位置。这些问题本质上是组合问题，因此找到最佳和/或接近最佳的决策在计算上是昂贵的。在本文中，提出了一个框架，通过首先使用深度卷积神经网络 (CNN) 预测与最佳位置相关的决策变量来降低底层整数数学程序的计算复杂度。CNN 以离线方式使用最佳解决方案进行训练，然后用于提供更小的优化问题，以进行实时决策。数值调查表明，所提出的方法可以以在线方式提供高质量的决策；一个对现实世界的实现至关重要的功能。