DP-MemArc: Differential Privacy Transfer Learning for Memory Efficient Language Models

Large language models have repeatedly shown outstanding performance across diverse applications. However, deploying these models can inadvertently risk user privacy. The significant memory demands during training pose a major challenge in terms of resource consumption. This substantial size places a heavy load on memory resources, raising considerable practical concerns. In this paper, we introduce DP-MemArc, a novel training framework aimed at reducing the memory costs of large language models while emphasizing the protection of user data privacy. DP-MemArc incorporates side network or reversible network designs to support a variety of differential privacy memory-efficient fine-tuning schemes. Our approach not only achieves about 2.5 times in memory optimization but also ensures robust privacy protection, keeping user data secure and confidential. Extensive experiments have demonstrated that DP-MemArc effectively provides differential privacy-efficient fine-tuning across different task scenarios.

Authors

• Yanming Liu Zhejiang University
• Xinyue Peng Southeast University
• Yuwei Zhang Tongji University
• Xiaolan Ke Harvard University
• Songhang Deng University of California, Los Angeles
• Jiannan Cao Massachusetts Institute of Technology
• Chen Ma Renmin University of China
• Mengchen Fu The University of Tokyo, Tokyo Institute of Technology
• Xuhong Zhang Zhejiang University
• Sheng Cheng Zhejiang University
• Xun Wang IIIS, Tsinghua University
• Jianwei Yin
• Tianyu Du

Keywords

No keywords are indexed for this paper.