Tabula: Efficiently Computing Nonlinear Activation Functions for Secure Neural Network Inference
Abstract: Multiparty computation approaches to secure neural network inference commonly rely on garbled circuits for securely executing nonlinear activation functions. However, garbled circuits require excessive communication between server and client, impose significant storage overheads, and incur large runtime penalties; for example, securely evaluating ResNet-32 using standard approaches requires more than 300MB of communication, over 10s of runtime, and around 5 GB of preprocessing storage. To reduce these costs, we propose an alternative to garbled circuits: Tabula, an algorithm based on secure lookup tables. Our approach precomputes lookup tables during an offline phase that contains the result of all possible nonlinear function calls. Because these tables incur exponential storage costs in the number of operands and the precision of the input values, we use quantization to reduce these storage costs to make this approach practical. This enables an online phase where securely computing the result of a nonlinear function requires just a single round of communication, with communication cost equal to twice the number of bits of the input to the nonlinear function. In practice our approach costs around 2 bytes of communication per nonlinear function call in the online phase. Compared to garbled circuits with quantized inputs, when computing individual nonlinear functions during the online phase, experiments show Tabula uses between $280 \times$-$560 \times$ less communication, is over $100\times$ faster, and uses a comparable amount of storage; compared against other state-of-the-art protocols Tabula achieves $>40\times$ communication reduction. This leads to significant performance gains over garbled circuits with quantized inputs during the online phase of secure inference of neural networks: Tabula reduces end-to-end inference communication by up to $9 \times$ and achieves an end-to-end inference speedup of up to $50 \times$, while imposing comparable storage and offline preprocessing costs.
Submission Length: Long submission (more than 12 pages of main content)
Assigned Action Editor: ~Antti_Honkela1
Submission Number: 1006
Loading