LOHEN: Layer-wise Optimizations for Neural Network Inferences over Encrypted Data with High Performance or Accuracy

Name: LOHEN: Layer-wise Optimizations for Neural Network Inferences over Encrypted Data with High Performance or Accuracy
Duration: 15 min
Description: Talk from 34th USENIX Security Symposium (USENIX Security '25).

Kevin Nam

34th USENIX Security Symposium (USENIX Security '25) · Day 3 · ML and AI Privacy 2