We propose the potential energy loss to make the forward embeddings more 'complicated', by pushing embeddings of the same class towards the decision boundary.

May 29, 2024 · In summary, PELoss provides significantly stronger privacy protection against model completion attacks, while preserving the model performance ...

Oct 18, 2022 · Abstract:As a practical privacy-preserving learning method, split learning has drawn much attention in academia and industry.

We propose the potential energy loss on the forward embeddings to push them to their decision boundary, in order to reduce the privacy leakage from the forward.

May 29, 2024 · The proposed potential energy loss method effectively reduces the performance of both fine-tuning attacks and clustering attacks, making it ...

Protecting Split Learning by Potential Energy Loss ... In this paper, we focus on the privacy leakage from the forward embeddings of split learning. Clustering ...

Sep 12, 2024 · As a privacy-preserving method for implementing Vertical Federated Learning, Split Learning has been extensively researched.

As a privacy-preserving method for implementing Vertical Federated Learning, Split Learning has been extensively researched. Privacy Preserving · regression +1.

It has been recognized that split learning is vulnerable to data privacy attacks, and various attacks along with corresponding defenses have been proposed.

A label-leakage attack that allows an adversarial input-owner to extract the private labels of the label-owner during split-learning, and evaluates the use ...