Presentation
An Enhanced Data Packing Method for General Matrix Multiplication in Brakerski/Fan-Vercauteren Scheme
DescriptionGeneral Matrix Multiplication (GEMM) is the most prevalent operation in deep learning (DL). However, performing GEMM within Fully Homomorphic Encryption (FHE) is inefficient due to high computational demands and significant data migration constrained by limited bandwidth. Additionally, the inherent limitations of FHE schemes restrict the widespread application of DL, as standard activation functions are incompatible. This incompatibility necessitates alternative nonlinear functions, which lead to notable accuracy reductions. To address these challenges, we propose a polynomial encoding method for GEMM under the Brakerski/Fan-Vercauteren (BFV) scheme and extend the method to inference with packing inputs and weights. Furthermore, we design specialized hardware to accelerate the inference process through optimized scheduling between the hardware and the host system. In experiments, we implemented our hardware on an FPGA U250 platform. Compared to existing solutions, our method achieves superior performance, achieving the highest 4.72 x and 3.78 x speedups on MNIST and CIFAR-10, respectively.
Event Type
Research Manuscript
TimeTuesday, June 2411:45am - 12:00pm PDT
Location3008, Level 3
SEC4: Embedded and Cross-Layer Security