Learning to Continuously Optimize Wireless Resource In Episodically Dynamic Environment

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.

References in corpus (26)

• Overcoming catastrophic forgetting in neural networks
• Playing Atari with Deep Reinforcement Learning
• A continual learning survey: Defying forgetting in classification tasks
• Recent Trends in Deep Learning Based Natural Language Processing
• Progressive Neural Networks
• Continual Learning Through Synaptic Intelligence
• Learning to Optimize: Training Deep Neural Networks for Wireless Resource Management
• Deep Learning-Based Communication Over the Air
• Deep Learning Methods for Improved Decoding of Linear Codes
• Learning to Detect
• Gradient Episodic Memory for Continual Learning
• AMP-Inspired Deep Networks for Sparse Linear Inverse Problems
• Experience Replay for Continual Learning
• Optimal Wireless Resource Allocation with Random Edge Graph Neural Networks
• Spatial Deep Learning for Wireless Scheduling
• Learning Optimal Resource Allocations in Wireless Systems
• Deep Unfolding: Model-Based Inspiration of Novel Deep Architectures
• Projection onto the probability simplex: An efficient algorithm with a simple proof, and an application
• Towards Robust Evaluations of Continual Learning
• DeepMIMO: A Generic Deep Learning Dataset for Millimeter Wave and Massive MIMO Applications
• Solving a Class of Non-Convex Min-Max Games Using Iterative First Order Methods
• Non-convex Min-Max Optimization: Applications, Challenges, and Recent Theoretical Advances
• Learning to Continuously Optimize Wireless Resource In Episodically Dynamic Environment
• Signal Processing and Optimal Resource Allocation for the Interference Channel
• Transfer Learning and Meta Learning Based Fast Downlink Beamforming Adaptation
• Learning to Beamform in Heterogeneous Massive MIMO Networks

Cited by in corpus (4)

• Learning to Continuously Optimize Wireless Resource in a Dynamic Environment: A Bilevel Optimization Perspective
• Learning to Continuously Optimize Wireless Resource In Episodically Dynamic Environment
• Algorithm Unrolling for Massive Access via Deep Neural Network with Theoretical Guarantee
• Edge Artificial Intelligence for 6G: Vision, Enabling Technologies, and Applications