FT-PrivacyScore: Personalized Privacy Scoring Service for Machine Learning Participation

TL;DR

This paper introduces FT-PrivacyScore, a prototype tool that provides personalized, quantitative privacy risk assessments for individuals participating in machine learning model fine-tuning, addressing a gap in privacy quantification.

Contribution

The paper presents a novel prototype that offers a practical method for individuals to estimate their privacy risk before engaging in machine learning tasks, filling a gap left by existing privacy measures.

Findings

Demonstrates the feasibility of quantitative privacy risk estimation

Provides a prototype tool for personalized privacy scoring

Enables informed decision-making for data contributors

Abstract

Training data privacy has been a top concern in AI modeling. While methods like differentiated private learning allow data contributors to quantify acceptable privacy loss, model utility is often significantly damaged. In practice, controlled data access remains a mainstream method for protecting data privacy in many industrial and research environments. In controlled data access, authorized model builders work in a restricted environment to access sensitive data, which can fully preserve data utility with reduced risk of data leak. However, unlike differential privacy, there is no quantitative measure for individual data contributors to tell their privacy risk before participating in a machine learning task. We developed the demo prototype FT-PrivacyScore to show that it's possible to efficiently and quantitatively estimate the privacy risk of participating in a model fine-tuning task. The demo source code will be available at \url{https://github.com/RhincodonE/demo_privacy_scoring}.