68 citations · 76 across the 2 of their papers we have counts for
4 papers
AESPA: Accuracy Preserving Low-degree Polynomial Activation for Fast Private Inference
Jaiyoung Park, Michael Jaemin Kim, Wonkyung Jung +1
Hybrid private inference (PI) protocol, which synergistically utilizes both multi-party computation (MPC) and homomorphic encryption, is one of the most prominent techniques for PI…
Accelerating Number Theoretic Transformations for Bootstrappable Homomorphic Encryption on GPUs
Sangpyo Kim, Wonkyung Jung, Jaiyoung Park +1
Homomorphic encryption (HE) draws huge attention as it provides a way of privacy-preserving computations on encrypted messages. Number Theoretic Transform (NTT), a specialized form…
Restructuring Batch Normalization to Accelerate CNN Training
Wonkyung Jung, Daejin Jung, and Byeongho Kim +3
Batch Normalization (BN) has become a core design block of modern Convolutional Neural Networks (CNNs). A typical modern CNN has a large number of BN layers in its lean and deep ar…
Partitioning Compute Units in CNN Acceleration for Statistical Memory Traffic Shaping
Daejin Jung, Sunjung Lee, Wonjong Rhee +1
The design complexity of CNNs has been steadily increasing to improve accuracy. To cope with the massive amount of computation needed for such complex CNNs, the latest solutions ut…