Kaiyuan Zhang (Purdue University), Siyuan Cheng (Purdue University), Guangyu Shen (Purdue University), Bruno Ribeiro (Purdue University), Shengwei An (Purdue University), Pin-Yu Chen (IBM Research AI), Xiangyu Zhang (Purdue University), Ninghui Li (Purdue University)

Federated learning collaboratively trains a neural network on a global server, where each local client receives the current global model weights and sends back parameter updates (gradients) based on its local private data.
The process of sending these model updates may leak client's private data information.
Existing gradient inversion attacks can exploit this vulnerability to recover private training instances from a client's gradient vectors. Recently, researchers have proposed advanced gradient inversion techniques that existing defenses struggle to handle effectively.
In this work, we present a novel defense tailored for large neural network models. Our defense capitalizes on the high dimensionality of the model parameters to perturb gradients within a textit{subspace orthogonal} to the original gradient. By leveraging cold posteriors over orthogonal subspaces, our defense implements a refined gradient update mechanism. This enables the selection of an optimal gradient that not only safeguards against gradient inversion attacks but also maintains model utility.
We conduct comprehensive experiments across three different datasets and evaluate our defense against various state-of-the-art attacks and defenses.

View More Papers

Explanation as a Watermark: Towards Harmless and Multi-bit Model...

Shuo Shao (Zhejiang University), Yiming Li (Zhejiang University), Hongwei Yao (Zhejiang University), Yiling He (Zhejiang University), Zhan Qin (Zhejiang University), Kui Ren (Zhejiang University)

Read More

The Midas Touch: Triggering the Capability of LLMs for...

Yi Yang (Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China; School of Cyber Security, University of Chinese Academy of Sciences, China), Jinghua Liu (Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China; School of Cyber Security, University of Chinese Academy of Sciences, China), Kai Chen (Institute of Information Engineering, Chinese Academy of…

Read More

A New PPML Paradigm for Quantized Models

Tianpei Lu (The State Key Laboratory of Blockchain and Data Security, Zhejiang University), Bingsheng Zhang (The State Key Laboratory of Blockchain and Data Security, Zhejiang University), Xiaoyuan Zhang (The State Key Laboratory of Blockchain and Data Security, Zhejiang University), Kui Ren (The State Key Laboratory of Blockchain and Data Security, Zhejiang University)

Read More