Reproduce “Communication-Learning Co-Design for Pinching-Antenna-Assisted Federated Learning in Heterogeneous Wireless Networks” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Shuai Wang, Wenxuan Sun, Mingjie Shao, Wantao Jiang, Yanqing Xu, Lingxiang Li, Zhi Chen
The authors propose a federated learning system that combines movable segmented waveguide antennas with control-variate updates and quantized model exchange to address wireless constraints and client drift. In CIFAR-10 image classification, under SWAN-OPT, it reaches 84.18% test accuracy versus 76.26% for FedAdam and reaches 70% in 81 communication rounds versus 225. These are simulation results, not an over-the-air trial.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.