Learning to Continuously Optimize Wireless Resource in Episodically Dynamic Environment
Learning to Continuously Optimize Wireless Resource in Episodically Dynamic Environment
复制标题
DOI：
10.1109/icassp39728.2021.9413503
复制
发表时间：
2020-11
期刊：
影响因子：
--
通讯作者：
Haoran Sun;Wenqiang Pu;Minghe Zhu;Xiao Fu;Tsung-Hui Chang;Mingyi Hong
中科院分区：
文献类型：
--
作者：
Haoran Sun;Wenqiang Pu;Minghe Zhu;Xiao Fu;Tsung-Hui Chang;Mingyi Hong
人们对开发数据驱动的、特别是基于深度神经网络 (DNN) 的现代通信任务方法越来越感兴趣。对于功率控制、波束成形和 MIMO 检测等一些常见任务，这些方法实现了最先进的性能，同时需要较少的计算量、较少的信道状态信息 (CSI) 等。然而，这些方法在 CSI 等参数不断变化的动态环境中学习通常具有挑战性。这项工作开发了一种方法，使数据驱动的方法能够在动态环境中持续学习和优化。具体来说，我们考虑“情景动态”设置，其中环境在“情景”中发生变化，并且在每个情景中环境都是固定的。我们提出了一个无线系统的持续学习（CL）框架，它可以逐步使学习模型适应新的场景，而不会忘记从之前的场景中学到的模型。我们的设计基于新颖的最小-最大公式，确保不同情节之间的一定“公平性”。最后，我们将 CL 方法定制为流行的基于 DNN 的功率控制模型，并使用合成数据和真实数据进行测试，从而证明了 CL 方法的有效性。
There has been a growing interest in developing data-driven, in particular deep neural network (DNN) based methods for modern communication tasks. For a few popular tasks such as power control, beamforming, and MIMO detection, these methods achieve state-of-the-art performance while requiring less computational efforts, less channel state information (CSI), etc. However, it is often challenging for these approaches to learn in a dynamic environment where parameters such as CSIs keep changing.This work develops a methodology that enables data-driven methods to continuously learn and optimize in a dynamic environment. Specifically, we consider an "episodically dynamic" setting where the environment changes in "episodes", and in each episode the environment is stationary. We propose a continual learning (CL) framework for wireless systems, which can incrementally adapt the learning models to the new episodes, without forgetting models learned from the previous episodes. Our design is based on a novel min-max formulation which ensures certain "fairness" across different episodes. Finally, we demonstrate the effectiveness of the CL approach by customizing it to a popular DNN based model for power control, and testing using both synthetic and real data.