Faster CryptoNets: Leveraging Sparsity for Real-World Encrypted Inference
This work addresses the problem of slow encrypted inference for privacy-preserving deep learning systems, making them more viable for real-world applications, though it is incremental as it builds on prior CryptoNets methods.
The paper tackles the computational inefficiency of homomorphic encryption for machine learning inference by introducing Faster CryptoNets, which uses pruning, quantization, and optimized activation approximations to achieve significant speedups while maintaining competitive accuracy.
Homomorphic encryption enables arbitrary computation over data while it remains encrypted. This privacy-preserving feature is attractive for machine learning, but requires significant computational time due to the large overhead of the encryption scheme. We present Faster CryptoNets, a method for efficient encrypted inference using neural networks. We develop a pruning and quantization approach that leverages sparse representations in the underlying cryptosystem to accelerate inference. We derive an optimal approximation for popular activation functions that achieves maximally-sparse encodings and minimizes approximation error. We also show how privacy-safe training techniques can be used to reduce the overhead of encrypted inference for real-world datasets by leveraging transfer learning and differential privacy. Our experiments show that our method maintains competitive accuracy and achieves a significant speedup over previous methods. This work increases the viability of deep learning systems that use homomorphic encryption to protect user privacy.