Robust Deep Learning for Wireless Network Optimization
Robust Deep Learning for Wireless Network Optimization
复制标题
DOI：
10.1109/icc40277.2020.9149445
复制
发表时间：
2020-06
期刊：
影响因子：
--
通讯作者：
Shuai Zhang;Bo Yin;Suyang Wang;Y. Cheng
中科院分区：
文献类型：
--
作者：
Shuai Zhang;Bo Yin;Suyang Wang;Y. Cheng
无线优化涉及反复解决困难的优化问题，并且数据驱动的深度学习技术有很大的希望，可以通过其模式匹配能力来减轻此问题：过去的最佳解决方案可以用作监督学习范式中的培训数据，以便神经网络可以使神经网络可以使用由于其高代表功率和并行实现，使用计算成本的一部分生成近似解决方案。但是，在网络方案中使这种方法实用需要仔细的，特定于领域的考虑，目前缺乏类似的作品。在本文中，我们在无线网络调度和路由中使用深度学习来预测是否将使用网络链接的子集，以减少有效的问题量表。现实世界中的关注是不同的数据重要性：由于类不平衡或标签质量不同，培训样本并不重要。为了弥补这一事实，我们开发了一种自适应样品加权方案，该方案在训练过程中动态加权批处理样本。此外，我们设计了一种新型损失功能，该功能使用其他网络层特征信息来提高解决方案质量。我们还讨论了一个后处理步骤，该步骤具有良好的阈值价值，以平衡预测质量和降低问题量表之间的权衡。通过数值模拟，我们证明了这些措施在从各种重要性数据中训练时，可以改善预测质量和规模的降低。
Wireless optimization involves repeatedly solving difficult optimization problems, and data-driven deep learning techniques have great promise to alleviate this issue through its pattern matching capability: past optimal solutions can be used as the training data in a supervised learning paradigm so that the neural network can generate an approximate solution using a fraction of the computational cost, due to its high representing power and parallel implementation. However, making this approach practical in networking scenarios requires careful, domain-specific consideration, currently lacking in similar works. In this paper, we use deep learning in a wireless network scheduling and routing to predict if subsets of the network links are going to be used, so that the effective problem scale is reduced. A real-world concern is the varying data importance: training samples are not equally important due to class imbalance or different label quality. To compensate for this fact, we develop an adaptive sample weighting scheme which dynamically weights the batch samples in the training process. In addition, we design a novel loss function that uses additional network-layer feature information to improve the solution quality. We also discuss a post-processing step that gives a good threshold value to balance the trade-off between prediction quality and problem scale reduction. By numerical simulations, we demonstrate that these measures improve both the prediction quality and scale reduction when training from data of varied importance.