GFlowNets for Active Learning Based Resource Allocation in Next Generation Wireless Networks
Charbel Bou Chaaya, Mehdi Bennis
TL;DR
This work tackles radio resource allocation in next-generation wireless networks that integrate communication, sensing, and computing. It introduces a generative active-learning framework based on Generative Flow Networks to sample resource-allocation matrices in high-dimensional discrete spaces, guided by a learned surrogate reward. A Gaussian-process surrogate coupled with trajectory-balanced GFlowNet training enables diverse, high-return solutions with fewer environment evaluations, achieving around 20% performance gains over strong baselines. The approach scales to larger device sets and offers a pathway to multi-fidelity extensions for further efficiency gains in dynamic wireless environments.
Abstract
In this work, we consider the radio resource allocation problem in a wireless system with various integrated functionalities, such as communication, sensing and computing. We design suitable resource management techniques that can simultaneously cater to those heterogeneous requirements, and scale appropriately with the high-dimensional and discrete nature of the problem. We propose a novel active learning framework where resource allocation patterns are drawn sequentially, evaluated in the environment, and then used to iteratively update a surrogate model of the environment. Our method leverages a generative flow network (GFlowNet) to sample favorable solutions, as such models are trained to generate compositional objects proportionally to their training reward, hence providing an appropriate coverage of its modes. As such, GFlowNet generates diverse and high return resource management designs that update the surrogate model and swiftly discover suitable solutions. We provide simulation results showing that our method can allocate radio resources achieving 20% performance gains against benchmarks, while requiring less than half of the number of acquisition rounds.