paper-with-me

Papers

On the Benefits of Public Representations for Private Transfer Learning under Distribution Shift

2023-12-24 · Pratiksha Thaker, Amrith Setlur, Zhiwei Steven Wu, Virginia Smith

Public pretraining is a promising approach to improve differentially private model training. However, recent work has noted that many positive research results studying this paradigm only consider in-distribution tasks, and may not apply to settings where there is distribution shift between the pretraining and finetuning data -- a scenario that is likely when finetuning private tasks due to the sensitive nature of the data. In this work, we show empirically across three tasks that even in settings with large distribution shift, where both zero-shot performance from public data and training from scratch with private data give unusably weak results, public features can in fact improve private training accuracy by up to 67\% over private training from scratch. We provide a theoretical explanation for this phenomenon, showing that if the public and private data share a low-dimensional representation, public representations can improve the sample complexity of private training even if it is impossible to learn the private task from the public data alone. Altogether, our results provide evidence that public data can indeed make private training practical in realistic settings of extreme distribution shift.

📄 PDF Abstract BibTeX arXiv:2312.15551

Code (0)

등록된 구현이 없습니다.

Tasks

regressionTransfer Learning

Similar Papers 제목 키워드 기반

Why Is Public Pretraining Necessary for Private Model Training?

2023-02-19 · Arun Ganesh, Mahdi Haghifam, Milad Nasr, Sewoong Oh 외

In the privacy-utility tradeoff of a model trained on benchmark language and vision tasks, remarkable improvements have been widely reported with the use of pretraining on publicly available data. This is in part due to …

Transfer Learning

Public policy for management of forest pests within an ownership mosaic

2023-12-08 · Andrew R. Tilman, Robert G. Haight

Urban forests provide ecosystem services that are public goods with local (shade) to global (carbon sequestration) benefits and occur on both public and private lands. Thus, incentives for private tree owners to invest i…

Management

Differentially Private Prototypes for Imbalanced Transfer Learning

2024-06-12 · Dariush Wahdany, Matthew Jagielski, Adam Dziedzic, Franziska Boenisch

Machine learning (ML) models have been shown to leak private information from their training datasets. Differential Privacy (DP), typically implemented through the differential private stochastic gradient descent algorit…

Transfer Learning

Public-private funding models in open source software development: A case study on scikit-learn

2024-04-09 · Cailean Osborne

Governments are increasingly funding open source software (OSS) development to support software security, digital sovereignty, and national competitiveness in science and innovation, amongst others. However, little is kn…

TransFR: Transferable Federated Recommendation with Pre-trained Language Models

2024-02-02 · Honglei Zhang, He Liu, Haoxuan Li, Yidong Li

Federated recommendations (FRs), facilitating multiple local clients to collectively learn a global model without disclosing user private data, have emerged as a prevalent architecture for privacy-preserving recommendati…

Privacy Preserving