Privacy-preserving learning for vehicle networks: applications and tools
This project is interdisciplinary of vehicle network, machine learning, and privacy preservation. We aim to address privacy concerns for the implementation of advanced functionalities on vehicles-especially in the scenarios of data sharing and exchanging-using methods of machine learning. We come up with applications that fit the goal, develop new algorithms to meet the requirement of learning efficiency and data privacy, exploit and extend open source tools to carry out both the expected algorithms and applications.
We have generated four applications, including energy-preserving route planning with privacy protection, privacy-aware mobility behavior classification, real-time traffic anomaly detection, privacy-preserving driving pattern analysis. We will choose at least one of them to be implemented with feasible algorithms that satisfies mentioned requirement, and carry them out using tools like TensorFlow Privacy, Diffprivlib, Pytorch Opacus, or other machine learning tools integrated with privacy-preserving mechanism. Verification will be conducted based on synthetically generated or real-world open source data to estimate the effectiveness of developed algorithms and applications.
Shiliang Zhang (contact)
Post doc at Chalmers, Computer Science and Engineering (Chalmers), Networks and Systems (Chalmers)
Swedish Research Council (VR)
Funding Chalmers participation during 2020–2021
Related Areas of Advance and Infrastructure
Information and Communication Technology
Areas of Advance
Areas of Advance
C3SE (Chalmers Centre for Computational Science and Engineering)
Innovation and entrepreneurship