Activation-Scaled ANN-to-SNN Conversion: A Universal Threshold Formula with Hybrid Architecture Insights

I present a simple yet effective formula for converting Artificial Neural Networks (ANNs) to Spiking Neural Networks (SNNs) with minimal accuracy loss. Key Findings (v2):- Universal threshold formula: θ = 2.0 × max(activation)- Tested on MLP, CNN, ResNet architectures- 100% accuracy preservation achieved (pseudo-MNIST)- Hybrid readout: 70% spike rate + 30% membrane potential- No training required - pure weight copy New in v2:- CIFAR-10 validation: formula remains stable on color images (64.2% SNN vs 68.2% ANN)- Hippocampus-inspired hybrid architecture: ANN feature extraction + SNN output achieves 100% accuracy preservation- Full SNN limitation discovered: converting all layers fails regardless of threshold tuning Recommended strategy: Keep feature extraction as ANN, convert only classification head to SNN with α=2.0. Code: https://github.com/hafufu-stack/temporal-coding-simulation/tree/main/ann-to-snn-converter