Learning to Continuously Optimize Wireless Resource In Episodically Dynamic Environment

TL;DR

This paper introduces a continual learning framework for deep neural network models to adapt to episodically dynamic wireless environments, enabling them to learn continuously without forgetting previous knowledge, thus maintaining high performance across changing scenarios.

Contribution

It develops a novel continual learning methodology with a min-max formulation for wireless systems, allowing models to adapt to episodic changes without losing prior knowledge.

Findings

Effective adaptation to new episodes with quick learning.

Maintains high performance on previous scenarios.

Validated on synthetic and real datasets.

Abstract

There has been a growing interest in developing data-driven and 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 to build the notion of continual learning (CL) into the modeling process of learning wireless systems, so that the learning model can incrementally adapt to the new episodes, {\it without forgetting} knowledge learned from the previous episodes. Our design is based on a novel min-max formulation which ensures certain ``fairness" across different data samples. We demonstrate the effectiveness of the CL approach by customizing it to two popular DNN based models (one for power control and one for beamforming), and testing using both synthetic and real data sets. These numerical results show that the proposed CL approach is not only able to adapt to the new scenarios quickly and seamlessly, but importantly, it maintains high performance over the previously encountered scenarios as well.