AI RESEARCH
ZC-Swish: Stabilizing Deep BN-Free Networks for Edge and Micro-Batch Applications
arXiv CS.LG
•
ArXi:2604.19453v1 Announce Type: new Batch Normalization (BN) is a cornerstone of deep learning, yet it fundamentally breaks down in micro-batch regimes (e.g., 3D medical imaging) and non-IID Federated Learning. Removing BN from deep architectures, however, often leads to catastrophic