RBOOT: Accelerating Homomorphic Neural Network Inference by Fusing ReLU within Bootstrapping

We present RBOOT, an novel and optimized framework that seamlessly integrates ReLU evaluation into CKKS bootstrapping, significantly reducing multiplication depth and boosting efficiency. By co-optimizing components of CKKS bootstrapping and novel approximation results of non-linear functions such as $\arcsin$, we can construct ReLU (and other non-linear functions) within the bootstrapping process itself, greatly reducing the computation overhead. Results on four widely used CNN models show that RBOOT achieves 2.77× faster end-to-end inference and 81% lower memory usage compared to previous polynomial approximation works, while maintaining comparable accuracy.