Machine Unlearning of Traffic State Estimation and Prediction

TL;DR

This paper introduces a machine unlearning approach for traffic state estimation and prediction models, enabling them to forget sensitive or outdated data to improve privacy and trustworthiness.

Contribution

It proposes a novel machine unlearning paradigm specifically designed for traffic state estimation and prediction models, addressing privacy and data freshness concerns.

Findings

Enables models to selectively forget privacy-sensitive data

Improves trustworthiness and reliability of TSEP models

Addresses privacy, cybersecurity, and data freshness issues

Abstract

Data-driven traffic state estimation and prediction (TSEP) relies heavily on data sources that contain sensitive information. While the abundance of data has fueled significant breakthroughs, particularly in machine learning-based methods, it also raises concerns regarding privacy, cybersecurity, and data freshness. These issues can erode public trust in intelligent transportation systems. Recently, regulations have introduced the "right to be forgotten", allowing users to request the removal of their private data from models. As machine learning models can remember old data, simply removing it from back-end databases is insufficient in such systems. To address these challenges, this study introduces a novel learning paradigm for TSEP-Machine Unlearning TSEP-which enables a trained TSEP model to selectively forget privacy-sensitive, poisoned, or outdated data. By empowering models to "unlearn," we aim to enhance the trustworthiness and reliability of data-driven traffic TSEP.