Reproduce “AS-FedBridge: Pseudo-Spike Bridge Distillation for Heterogeneous ANN-SNN Federated Learning” 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
Shengyang Li, Yiting Dong, Liuyang Song, Ximing Wang, Luyuan Xie, Cong Li, Qingni Shen, Zhaofei Yu
AS-FedBridge enables federated learning between clients using conventional artificial neural networks and spiking neural networks without sharing their private backbones. It uses a shared Bridge with a pseudo-spike interface and bidirectional knowledge transfer, aligning an exchange representation with SNN firing rates. On CIFAR-100, it reaches 71.64% mean client accuracy versus 68.75% for FML under the tested ten-client, non-IID setup.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.