Characterizing and Optimizing End-to-End Systems for Private Inference

Garimella, Karthik; Ghodsi, Zahra; Jha, Nandan Kumar; Garg, Siddharth; Reagen, Brandon

doi:10.1145/3582016.3582065

Cited by 6 publications

(2 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HE introduces a large computational overhead of 4-6 orders of magnitude slowdown over their plaintext counterparts (1080 seconds for a single ResNet-18 inference [10,38]). Rather than directly using HE for computing linear layers, it is common to combine HE with Additive Secret Sharing (SS), another cryptographic building block that supports plaintext-level speeds for computing linear layers.…”

Section: Private Linear Computation (He Ss)mentioning

confidence: 99%

“…The combination of the proposed optimizations improves the mean inference latency by 1.8× over the state of the art. [10] Besides the system-level optimizations, we also review two hardware accelerators, HAAC [30] and RPU [40], to accelerate GCs and HE, respectively. HAAC aims at reducing the computation latency for the garbled circuits generation/evaluation, while RPU targets the offline HE computation overhead.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Towards Fast and Scalable Private Inference

Garimella

Neda

et al. 2023

Proceedings of the 20th ACM International Conference on Computing Frontiers

Self Cite

View full text Add to dashboard Cite

Privacy and security have rapidly emerged as first order design constraints. Users now demand more protection over who can see their data (confidentiality) as well as how it is used (control). Here, existing cryptographic techniques for security fall short: they secure data when stored or communicated but must decrypt it for computation. Fortunately, a new paradigm of computing exists, which we refer to as privacy-preserving computation (PPC). Emerging PPC technologies can be leveraged for secure outsourced computation or to enable two parties to compute without revealing either users' secret data. Despite their phenomenal potential to revolutionize user protection in the digital age, the realization has been limited due to exorbitant computational, communication, and storage overheads.This paper reviews recent efforts on addressing various PPC overheads using private inference (PI) in neural network as a motivating application. First, the problem and various technologies, including homomorphic encryption (HE), secret sharing (SS), garbled circuits (GCs), and oblivious transfer (OT), are introduced. Next, a characterization of their overheads when used to implement PI is covered. The characterization motivates the need for both GCs and HE accelerators. Then two solutions are presented: HAAC for accelerating GCs and RPU for accelerating HE. To conclude, results and effects are shown with a discussion on what future work is needed to overcome the remaining overheads of PI. CCS CONCEPTS• Security and privacy → Privacy-preserving protocols; • Computer systems organization → Architectures.

show abstract

Section: Private Linear Computation (He Ss)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%